Does early speaking lead to fossilized mistakes?

[s_allard]

The whole premise of adult immersion classes such as that of Middlebury College in the US and in countless schools across the world is to expose the student, receptively and productively, to the language as early as possible.

Mistakes when learning a language are a way of life. Whether these mistakes are old bad habits, i.e. fossilized, or recent ones, there is no difference. The problem is a) how to detect them and b) how to eventually eliminate them.

The big problem with language mistakes is that we usually don't know we are making them. And that's the whole point. Unlike the native speaker who usually has a good knowledge of what is correct speech and is able to immediately correct a slip of the tongue, as learners we do not have perfect knowledge of what is proper speech.

We are therefore usually unaware of a mistake. The first step then is to have the mistake corrected or drawn to our attention. This is a huge problem for adults. If no one corrects you, you are screwed. This is why a tutor is so important if you do not have free correction from spouses, lovers, friends, teachers, etc.

The second step is to change one's linguistic behaviour once a mistake has been brought to our attention. This is the hard part and can take various forms. What I recommend is deliberate and conscious repetitive use of the corrected form as soon as possible.

So, let's say I'm having a problem with the pronunciation of Spanish independientemente because of all the syllables. I simply make of point of using it a few times every day for the next few days until I feel I have it.

Grammar and vocabulary mistakes are harder to detect because they may not be noticeable unless somebody calls them to my attention. Also at more advanced levels, the learners will have some doubt. "Hmm, is that word masculine or feminine?" "Is this the right preposition?"

Once the mistake is brought to your attention, then you deal with it in the appropriate manner.

The big problem in all this is error correction. Unless you are corrected, you will become fossilized.

[Cavesa]

I don't think the problem is early or late speaking per se. The problem is speaking without getting the theoretical background, especially grammar.

I am all for early speaking at least to oneself, all for acquiring the pronunciation first. But the problem is skipping the rest of the process. Only some of the experienced learners can learn just from native sources without any coursebooks, we have awesome examples in our community. But most learners trying it are not the case. I personally wouldn't attempt it, and I consider myself a much more experienced person in the matter than the usual first time monolingual learner asking for the best apps and exchange sites on reddit.

There are many learners who simply get more and more "fluent" at neanderthalian version of the given language. And no correction in the world is gonna make up for skipping a good grammar explanation. I've heard teachers correcting the same mistakes over and over and over in class. But it is an absolute waste of time, unless the student opens the book and spends time on the matter. Correction on its own doesn't work. It may make the person memorize the particular example but it doesn't treat the systematic underlying problem.

Even worse is the very popular "you can learn to speak only by speaking" and "just do the reverse tree and speak!" and "grammarbooks are useless, just learn a phrasebook and speak from day one" attitude. Yes, you can learn to speak correctly only by immersion, if you are 100% immersed, your life depends on it, and you have a certain kind of intelligence (I am not saying a certain IQ, not at all! I am talking about the fact there are various kinds of intelligence and some people do learn a language to perfection after moving abroad and some don't even after decades of honest trying). I couldn't do it, I will never fulfill all these conditions. I would probably become one of the fluent neanderthals.

Most learners attempting it can't, they just take the part "throw away boring grammarbooks and courses" and practice the mistakes. That is the problem. And after certain point, removal of the mistakes is extremely hard.

[Random Review]

OK this may seem like I'm going off-topic, because obviously these people are not speaking from day 1, but please bear with me and I hope you will see the relevance.

As it happens I know literally hundreds of people (and have met many thousands more) who had really bad "fossilised" errors when learning a language and yet went on to speak it perfectly: they are called native speakers. I have even watched this process happen with my youngest sister (10 years younger than me) and my two nieces. In fact one of my nieces is trilingual (her mum is British, dad Romanian and she lived in Spain all her life until a couple of months ago, so I got to see her do this in two languages- sadly I don't speak Romanian). I watched her go from telling me (I'm a guy) that I was "castigada" (for those who don't know Spanish, that's the feminine form of the adjective) to speaking much better Spanish than me (as a 9-year-old native) just a few years later.

As something that we all know, this ought to raise more interest than it does IMO.

My own thought is that if you learn mechanically, errors will fossilise; if you don't, they won't. To stretch an analogy, if you build a house on faulty foundations, you can't build very high; but a living tree can start off wonky as a sapling and grow into the tallest, straightest thing you ever saw. Sadly most other-directed (i.e. teachers, courses, etc) learning is mechanical, which is why I am currently avoiding speaking Mandarin as much as I possibly can (I'm in China right now): I'm at that crucial stage of laying foundations (tones, prosody, difficult consonants, etc).

I don't pretend to understand this fully, but I think you can get a rough idea of how this works: tones, phonemes, etc are building blocks I am going to combine with basic grammar into simple sentences, which in turn I am going to use with more complex grammar to express more complex ideas, which I am then going to use with a (hopefully) growing vocabulary to express my thoughts in Mandarin.

This is mechanical, it's building. Children don't learn like this, nor do successful speak-from-day-one polyglots IMO.

When you start to think like this, all sorts of strange things about the strategies independent learners use start to make sense: why is it so important to use books you love for LR to work fully? Why do the best Assimil courses waste space with that weird humour, quirky characters and teach vocabulary that can express quirks of personality and not more practical stuff? Why would the 'lyrics training' site be any better than a dictation exercise in a class? Why does my (not so good) Spanish accent improve for a short time when I listen to several hours of music in the language? Why does it improve when I talk to someone I feel a connection with? Why did I once have a student (back when I taught adults in Spain) who spoke with a strong Spanish accent but could sing with an extremely good accent? Actually, I saw something similar to this my first week in China. One of the weakest students I have here in China is a little girl of maybe 6 or 7 whose English in class is horrible. The thing is, I observed this class before I taught it and heard her sing (I assume unbeknownst to the teacher who was teaching the class, because he never tried to get her attention back onto his class) 'Edelweiss' very, very well indeed. Then there is the case of my very youngest students and the word "phonics" that I mentioned in another thread (I will find the link and post it here).

Speaking of teaching, I think this is the grain of truth behind the fetish for student-student interactions in the EFL industry. If you take a basic (and probably the most common) methodology like PPP (present-practice-produce) for instance, then for the practice stage I totally disagree with the industry in general* and agree with the small minority who say that we should maximise teacher-student interactions: this stage is mechanical and usually they just learn each other's errors and substandard pronunciation (and for the exact same reason, I think there should be much more error correction at this stage); on the other hand the 'produce' stage (if it really is this and the lesson plan isn't just paying lip service to the idea) should, by definition be meaningful and communicative and to achieve this, you really do need to maximise student-student interactions, minimise teacher involvement and ignore a lot of errors so as to facilitate this.

So how to use this to avoid fossilised errors? Well, if I knew that I wouldn't be following the strategy that I am with Mandarin! One thing I will confidently state, though: it should involve joy and meaningful interaction with real people (this can include reading the books of dead authors or watching the movies of directors and screenwriters who will never know you exist). If it is not fantastic fun and if no real communication is taking place, it is mechanical learning at best (or useless at worst). That is why the notion of "input" in popularisations of Krashen are useful if taken intelligently as a way to quantify what you are doing but damaging if taken too literally: input is such a mechanical concept (whereas Krashen's idea about acquiring a language by understanding messages is about communication, which is a living thing usually involving back and forth and a message that genuinely interests you).

Sorry for the poor organisation of this post, it reflects the lack of organisation of my thoughts on this, as it's something I'm still exploring.

* If anyone thinks it's a bit arrogant of me to say I disagree with people who (I acknowledge) are far better and more experienced teachers than me, I do it as someone who has actually learned a second language as an adult to a decent level, which very few EFL teachers have done in my experience. Off topic a little, there's a guy I mentioned in another thread who claims to speak Spanish and whose Spanish is horrendous. He very kindly offered to advise me about learning Mandarin. I declined as politely as I could.

Edit: @ Cavesa's post above: I think one of the differences between the experienced learners who do that successfully and most people is that they are not learning mechanically in the way I talk about above.

[s_allard]

When we speak of fossilized mistakes, as is the case here, what we are usually talking about are old bad habits that we keep making despite all our studying of the language including the very mistakes in question. A classic example is the English-speaking learner of French who keeps saying "beaucoup de temps" instead of "beaucoup de fois" for "many times".

This person has seen usage of the words temps and fois hundreds of times and perfectly understands the difference. But why does this person keep making the same mistake? The problem is the process of error correction. Because this error is so ingrained it must be dealt with specifically. A quick ad hoc correction won't do. Deliberate practice and use for a few days should do the trick.

[Random Review]

When we speak of fossilized mistakes, as is the case here, what we are usually talking about are old bad habits that we keep making despite all our studying of the language including the very mistakes in question. A classic example is the English-speaking learner of French who keeps saying "beaucoup de temps" instead of "beaucoup de fois" for "many times".

This person has seen usage of the words temps and fois hundreds of times and perfectly understands the difference. But why does this person keep making the same mistake? The problem is the process of error correction. Because this error is so ingrained it must be dealt with specifically. A quick ad hoc correction won't do. Deliberate practice and use for a few days should do the trick.

I don't know if the word "perfectly" is really right there, mate. Maybe the key is that they don't understand it perfectly, with enough connections to other knowledge and memories and real life situations. Maybe they only understand it in an abstract sense.

Yesterday I had an embarrassing experience in a coffee shop in China: I ordered a latte as usual and was asked a question involving the strange word rède. Eventually they found someone who spoke English and it turned out that they were asking me if I wanted it hot ("rè de" and no, they weren't offering me an iced latte). Thing is, I know the word for "hot" (rè) and I know that exact grammatical construction with adjective + "de". I reviewed this expression not 3 days ago in Pimsleur and I could even have translated that exact sentence from English to Mandarin. Do I know the word? Kind of. I know what it means, I know how to pronounce it, I can recognise it in a context and produce it from an English prompt.

I clearly don't know it perfectly though, because the very fact that it never occurred to me that anyone would ever ask me if I wanted my latte hot (that was the first time in my life I have ever been asked that) was sufficient to throw me enough that I didn't recognise it.

[s_allard]

When we speak of fossilized mistakes, as is the case here, what we are usually talking about are old bad habits that we keep making despite all our studying of the language including the very mistakes in question. A classic example is the English-speaking learner of French who keeps saying "beaucoup de temps" instead of "beaucoup de fois" for "many times".

This person has seen usage of the words temps and fois hundreds of times and perfectly understands the difference. But why does this person keep making the same mistake? The problem is the process of error correction. Because this error is so ingrained it must be dealt with specifically. A quick ad hoc correction won't do. Deliberate practice and use for a few days should do the trick.

I don't know if the word "perfectly" is really right there, mate. Maybe the key is that they don't understand it perfectly, with enough connections to other knowledge and memories and real life situations. Maybe they only understand it in an abstract sense.

...

The point is well taken. We could assume that if people knew things well enough they wouldn't make mistakes. In the particular case I was referring to the person understanding the difference between the two words in this context. One could argue that maybe the person never perfectly understood the use of the individual words in general. It is something to think about.

[Random Review]

When we speak of fossilized mistakes, as is the case here, what we are usually talking about are old bad habits that we keep making despite all our studying of the language including the very mistakes in question. A classic example is the English-speaking learner of French who keeps saying "beaucoup de temps" instead of "beaucoup de fois" for "many times".

This person has seen usage of the words temps and fois hundreds of times and perfectly understands the difference. But why does this person keep making the same mistake? The problem is the process of error correction. Because this error is so ingrained it must be dealt with specifically. A quick ad hoc correction won't do. Deliberate practice and use for a few days should do the trick.

I don't know if the word "perfectly" is really right there, mate. Maybe the key is that they don't understand it perfectly, with enough connections to other knowledge and memories and real life situations. Maybe they only understand it in an abstract sense.

...

The point is well taken. We could assume that if people knew things well enough they wouldn't make mistakes. In the particular case I was referring to the person understanding the difference between the two words in this context. One could argue that maybe the person never perfectly understood the use of the individual words in general. It is something to think about.

If you can permit me to hazard a guess, I'd bet on most of these people understanding "fois" and the difference between "fois" and "temps" but unconsciously having a faulty understanding of "temps" as being equivalent to "time".

One thing I realised when I would make mistakes with the subjunctive in Spanish (I assume the case in French is similar) is that I would make two sorts of mistake. Occasionally I would use the wrong mood because I didn't understand the distinction well enough; but mostly I would make mistakes where I would use the indicative in a place where I knew fine well that I should use the subjunctive. The problem was that for all I had studied more and more details of when the subjunctive is correct, I had never updated my mistaken mental model of 'Spanish indicative = English indicative'. What helped me with this was the work of a guy called Jose Ruiz Campillo, which makes clear the exact nature of the Spanish indicative.

If the fois/temps distinction in French is the same as the vez/tiempo distinction in Spanish, then I'd love for you to try the following experiment and tell me if it works. Get them to draw a representation with dots of the French for "I bought it 3 times" (or whatever) and then a representation with a line of the French for "I lived there for 3 years" (or whatever), if my idea is right, they should be able to do this no problem; then ask them to draw a representation of the incorrect sentence they just said with "temps" and hopefully enjoy watching them start to draw, hesitate and then the lightbulb going off.

In other words and following on from my long post above, I guess I'm saying that I think that these people have mostly thought about the meaning of "fois" (because it will have surprised them that French has a separate word for that) and also about when to use "fois" and when "temps" (they have probably even practiced this in a mechanical way*), but never given any thought to what "temps" means in the light of this (this is not illogical, in our own native languages we often learn a more precise term and when to use it when studying something, but continue using the broader term as well in our everyday speech).

* because to do this kind of exercise, you only need to truly understand one of the pair and can apply the other one mechanically anywhere else without having to pay attention to its meaning.

[blaurebell]

The whole premise of adult immersion classes such as that of Middlebury College in the US and in countless schools across the world is to expose the student, receptively and productively, to the language as early as possible.

I have decided to stay largely out of this discussion from now on, since clearly none of what I wrote in the other thread, where this topic came up before, has made any impression on you. In fact, you spoke about me in the third person again, although I made it very clear that I have a strong dislike for that. I could clearly spend my time better by talking to a wall.

[s_allard]

When we speak of fossilized mistakes, as is the case here, what we are usually talking about are old bad habits that we keep making despite all our studying of the language including the very mistakes in question. A classic example is the English-speaking learner of French who keeps saying "beaucoup de temps" instead of "beaucoup de fois" for "many times".

This person has seen usage of the words temps and fois hundreds of times and perfectly understands the difference. But why does this person keep making the same mistake? The problem is the process of error correction. Because this error is so ingrained it must be dealt with specifically. A quick ad hoc correction won't do. Deliberate practice and use for a few days should do the trick.

I don't know if the word "perfectly" is really right there, mate. Maybe the key is that they don't understand it perfectly, with enough connections to other knowledge and memories and real life situations. Maybe they only understand it in an abstract sense.

...

The point is well taken. We could assume that if people knew things well enough they wouldn't make mistakes. In the particular case I was referring to the person understanding the difference between the two words in this context. One could argue that maybe the person never perfectly understood the use of the individual words in general. It is something to think about.

If you can permit me to hazard a guess, I'd bet on most of these people understanding "fois" and the difference between "fois" and "temps" but unconsciously having a faulty understanding of "temps" as being equivalent to "time".

One thing I realised when I would make mistakes with the subjunctive in Spanish (I assume the case in French is similar) is that I would make two sorts of mistake. Occasionally I would use the wrong mood because I didn't understand the distinction well enough; but mostly I would make mistakes where I would use the indicative in a place where I knew fine well that I should use the subjunctive. The problem was that for all I had studied more and more details of when the subjunctive is correct, I had never updated my mistaken mental model of 'Spanish indicative = English indicative'. What helped me with this was the work of a guy called Jose Ruiz Campillo, which makes clear the exact nature of the Spanish indicative....

If I understand correctly, here we are talking about two kinds of mistakes of lexical usage. One is not understanding the distinction between two (or more terms). The other is associating one term erroneously with a similar term in the native language. For example, we could assume that the person thinks "temps" is closer to "time" because of the resemblance. Similarly, in Spanish I tended to say "exprimirse" instead of "explicarse" because of the resemblance to the French "s'exprimer".

The correction of my fossilized Spanish mistakes basically involves a) learning how explicarse works and b) not to use exprimirse when I want the equivalent of the French s'exprimer and possibly c) learn how exprimirse works in Spanish so that I won't mix it up with the French lookalike.

To come back to the theme of the thread, I think that the advantages of early exposure and even early speaking far outweigh the disadvantages. Sure, some people can plateau at a certain level for a variety of reasons. Progress comes from constant and proper acquisition plus error correction. If you stop, then you plateau.

There is a particular problem for the advanced learners, people who are in the B2/C1 range and want to go higher. I know this can be particularly frustrating because there is not much material aimed at the really advanced learners. We're heading here into the zone of close to native performance.

That's exactly why I suggested that instead of complaining about being stuck at B2, a great learning experience would be to take actual classes in the language but not language classes per se. This is the ultimate challenge and probably requires help. And here more than ever the two principles of constant acquisition and constant error correction still apply.

[qeadz]

While I cannot add to this discussion in a meaningful way because I have never studied how languages are acquired, I do like to pit assertions against my own experience and observations.

When I think back to my early school days - the bits that I can remember (which was all pre-internet so the status updates, facebook feeds, even text messaging echo chambers did not exist) - native speakers learning their own language still had to go through numerous corrections (both grammar and spelling) to mistakes which had become fossilized through years of misunderstanding.

We would keep making the same mistakes every English class and it was like whack-a-mole. Every thing we wrote in English class saw recurring issues.

Most kids had by end of high school made significant improvements. Some kids who once struggled along with their classmates in correctly applying simple grammar went on to be incredibly well-spoken in their adult life.

So at this juncture I have decided not to worry about fossilization for two reasons:
1) I seem to be capable of overcoming it - at any stage of learning (fun trivia: until my mid 20's I actually had never known when to use 'less' and when to use 'fewer'. I now use them appropriately when speaking)
2) For each mistake that 'fossilizes', I believe I reinforce many more desirable connections and there is a significant net gain despite the collateral fossilization.