Chunking...

[jeffers]

When I read the description of the "Lexical Approach", I thought it sounded familiar. I was studying theology in the late 90s, and a couple of my professors of hermeneutics wrote and spoke a lot about the "structural approach". Here's my rough and ready summary of their approach. All my own summary, from my vague memory dating back to the 90s:

The oft repeated phrase from both of these professors was that "the basic unit of language is not the word, but the phrase". In other words, traditional language learning (and Biblical hermeneutics) focused on learning a bunch of words and learning grammar to know how to fit these words together, but in fact speakers think in whole phrases, with maybe swappable words which can be passed in and out. "I want some [water]" is a fixed standard phrase, and children learn it as a whole. While mastering such basic phrases childrean also learn to play with them, and it is simple enough to swap some other need for [water], or add in some sort of negative. As they play and make mistakes, they eventually morph the fixed good phrase into another good phrase: "I [no] want some [sandwich]" --> "I don't want a sandwich". Traditionally, hermeneutics textbooks loved the study of etymology, but these professors also emphasized that the meaning of a word derives from its context, and from contemporary usage far more than the history of the word. Example, most people who use "glamour" don't have a clue that it means "magic". This might seem to be a slight digression from the structural approach, but Moisés Silva applied the structural approach to the study of words in his 1983 book, Biblical Words and their Meaning.

These professors did not discourage studying a lot of vocabulary and grammar, but their use of the structural approach came into play more when studying a text (because, of course, they were teaching hermeneutics, i.e. interpretation). They encouraged us to read whole sections first, look for the larger meaning of the whole, before drilling down to asking why a specific word is used here instead of another, or why this particular tense was chosen over another. And they strongly implied that spending time reading the biblical texts was more useful for learning the languages than spending time flipping vocabulary cards, given a basic core of knowledge from first year Greek or Hebrew.

This structural approach sounds a lot like the "Lexical Approach", except that the structural approach is much older, dating back to the work of the Swiss linguist Ferdinand de Saussure (according to my search in Wikipedia just now), who died in 1913. So, add to what Emk wrote about common patterns in education the fact that the "brilliant professor" who "gets good results" has probably picked up the ideas from somewhere but doesn't know where, so then their followers begin to attribute the "new" approach to that professor's genius and a movement is born.

[Slowpoke]

Unrelated, but I was excited to talk about Wong Kar Wai when I saw the title.

[miket12]

lOne of my friends is an extreme example: She abandoned her native language abruptly at age 5, because nobody at school spoke it. But her mother kept speaking it to her. Today, she can watch gritty cop dramas in her original native language, and even read her family's Facebook posts. But she swears up and down that she can barely do A1-level speaking tasks.

I can give a similar story. My mother grew up in the era when immigrants wanted their children to be what they though of as American so they forbade her from speaking their language, but her parents still spoke it to each other. My mom always used to say that she could understand without any problems but was unable to speak it.

While I'm not a neuroscientist I've read that there are two distinct brain areas involved. Talking takes place in Broca’s area (among others) but understanding speech happens in Wernicke’s area (among others). A stroke patient, depending on where the stroke happens, may have a deficit in one or the other. And it seems plausible that when learning a language if you listen but don't speak Wernicke’s area gets trained but Broca's does not.

[Nikita]

I find this topic really interesting because it describes the method I have been using for a long time in my language learning. Initially I started memorizing phrases that are commonly used in English but don't make sense in my native language. My idea was to incorporate them into my writing to make it more natural. However, I realized that it allows me to write naturally sounding sentences without translating from Russian, just by combining and slightly altering the phrases I learnt before. After that I decided to memorize sentences looking potentially applicable in different contexts regardless of whether they are peculiar to English or actually similar to Russian. Following this approach helped me to gain the confidence I needed to express myself in English and also allowed me to stop translating from my native language.

I continue practicing my active language skills this way because it enables me to completely avoid mistakes caused by translating directly from Russian and I also find it more exciting than studying grammar rules (although I keep reviewing them every now and then). This approach also allows me to create impressively sophisticated sentences I would never come up with even in my native language, although with the advent of ChatGPT it has ceased to evoke unfeigned astonishment. The idea of memorizing phrases to use them later in my writing and speaking runs through all my language learning process, and you can probably notice it even in this comment.

I think It would be a wrong oversimplification to believe that you can obtain fluency only by memorizing preconstructed phrases. The way I see it is that learning grammar is necessary to be able to create meaningful sentences by connecting different chunks and making a smooth transition between them. Without having a good grasp of grammar constructions, your ability to communicate your ideas will be too limited and your speaking will lack the variety that people normally expect from an advanced language learner.

Despite my long-standing appreciation for this method, I don't consider it to be the most effective way to enhance language skills. My improvement should be attributed to consistent practice rather than to the effectiveness of chunking. I am also aware that my English writing is neither grammatically perfect nor completely natural (in the sense that it is quite different from how most native English speakers usually express themselves). In my opinion, chunking is the most suitable for academic writing where having a huge set of very specific and deliberately memorized phrases appears to be a precondition for being able to maintain a formal voice through the whole article. However, this technique is less useful in informal speaking where spontaneity is appreciated more than accuracy.

The main issue seems to be that some teachers tend to present this method as the best approach to attain proficiency in a foreign language and even encourage students to abandon all other language acquisition techniques as less effective. Chunking is excellent for improving your active vocabulary and overall confidence when using your target language, but you should not disregard the importance of grammar training.

[Cainntear]

I think It would be a wrong oversimplification to believe that you can obtain fluency only by memorizing preconstructed phrases. The way I see it is that learning grammar is necessary to be able to create meaningful sentences by connecting different chunks and making a smooth transition between them. Without having a good grasp of grammar constructions, your ability to communicate your ideas will be too limited and your speaking will lack the variety that people normally expect from an advanced language learner.

Despite my long-standing appreciation for this method, I don't consider it to be the most effective way to enhance language skills. My improvement should be attributed to consistent practice rather than to the effectiveness of chunking. I am also aware that my English writing is neither grammatically perfect nor completely natural (in the sense that it is quite different from how most native English speakers usually express themselves). In my opinion, chunking is the most suitable for academic writing where having a huge set of very specific and deliberately memorized phrases appears to be a precondition for being able to maintain a formal voice through the whole article. However, this technique is less useful in informal speaking where spontaneity is appreciated more than accuracy.

Well not... chunking is simply a part of how we process language: chunks made of chunks made of chunks. Chunks all the way down.

The main issue seems to be that some teachers tend to present this method as the best approach to attain proficiency in a foreign language and even encourage students to abandon all other language acquisition techniques as less effective. Chunking is excellent for improving your active vocabulary and overall confidence when using your target language, but you should not disregard the importance of grammar training.

No, the main issue is that some teachers present their method as being informed by the psychological construct of "chunks", a concept which they don't quite understand.

The error is common in lots of fields: new science comes out and people rush to do something that they don't really understand.
In education and psychology in general, it's often a matter of looking at a base principle and then trying to build something around that, ignoring everything over the base level initially.

For example, behaviorism suggests psychology is just reaction to stimulus. It may well be, but the stimulus is so complex that we can't predict the reaction, so it's entirely academic. That didn't stop behaviorists acting as though they could. It also didn't stop teachers trying to revise their models "informed by the latest psychology"... and we got a new type of rote learning where they thought that answering the same questions the same way would build "habits" in reaction to stimulus. Magical thinking... didn't work.

Chunking was picked up by te teaching world early on and misinterpreted. They used it to justify teaching specific chunks, but the chunks are often very subtle and the most important ones weren't always noticed by the teachers....

[Sprachprofi]

Thanks for extracting this into a thread, Cainntear!

Just for the record: I'm not anti-grammar, I just thought that this particular person, with high comprehension and zero speaking, would have little use of it. Imho grammar is a crutch on the way to developing a feeling for what "sounds right". Once you have that feeling, grammar is more likely to do harm than good, because its rules never cover all cases.

[Nikita]

No, the main issue is that some teachers present their method as being informed by the psychological construct of "chunks", a concept which they don't quite understand.

The error is common in lots of fields: new science comes out and people rush to do something that they don't really understand.
In education and psychology in general, it's often a matter of looking at a base principle and then trying to build something around that, ignoring everything over the base level initially.

For example, behaviorism suggests psychology is just reaction to stimulus. It may well be, but the stimulus is so complex that we can't predict the reaction, so it's entirely academic. That didn't stop behaviorists acting as though they could. It also didn't stop teachers trying to revise their models "informed by the latest psychology"... and we got a new type of rote learning where they thought that answering the same questions the same way would build "habits" in reaction to stimulus. Magical thinking... didn't work.

Chunking was picked up by te teaching world early on and misinterpreted. They used it to justify teaching specific chunks, but the chunks are often very subtle and the most important ones weren't always noticed by the teachers....

I think your extensive familiarity with both linguistics and psychology makes you perceive it differently from most people. You look deeply into the essence of how we learn languages, whereas teachers usually only seek to provide their students with useful techniques that facilitate language acquisition. Focusing on memorizing short phrases instead of separate words has been proven to be effective and that is why this method has gained popularity. From the language learners' perspective it doesn't seem to matter whether the term "chunking" is used exactly as it is understood in psychology or not. You said that some chunks are very subtle and habitually overlooked by most teachers, but I don't know what you really mean. Maybe my understanding is too superficial and I also fail to notice them.

What I was trying to get across is that from a purely practical standpoint, this approach brings better results where you can clearly see which phrases you should memorize to be able to produce natural sentences. In my experience, texts written in formal English are easily analysable into chunks, but in speaking, where native speakers express themselves in a less structured way and use the grammar that deviates starkly from textbook English, patterns are much more difficult to identify. So, the extent to which making a conscious effort to extract these patterns from the input you are exposed to helps you to improve may vary. I agree that regardless of your approach, in the long run you will end up thinking in chunks and not separate words just because this is how your brain processes language in general.

[Cainntear]

I think your extensive familiarity with both linguistics and psychology makes you perceive it differently from most people.

Yes -- I percieve it correctly.

You look deeply into the essence of how we learn languages, whereas teachers usually only seek to provide their students with useful techniques that facilitate language acquisition.

Ah, so does that mean I'm not really a teacher then...?

Focusing on memorizing short phrases instead of separate words has been proven to be effective and that is why this method has gained popularity. From the language learners' perspective it doesn't seem to matter whether the term "chunking" is used exactly as it is understood in psychology or not. You said that some chunks are very subtle and habitually overlooked by most teachers, but I don't know what you really mean. Maybe my understanding is too superficial and I also fail to notice them.

But I'm not simply saying that teachers use the term differently from psychology -- I'm saying that teachers have picked up the term from psychology and keep saying that the psychology supports what they're doing, when it really doesn't.

You are absolutely right that chunking as defined by psychology isn't really something for the learner to worry about. My problem is phrase-based methodology doesn't teach chunking anyway, but chunking is used as a justification or rationalisation of the approach.

What I was trying to get across is that from a purely practical standpoint, this approach brings better results where you can clearly see which phrases you should memorize to be able to produce natural sentences.

Well... natural looking sentences. But are they truly natural sentences if the learner has not gone through the natural process to produce it...?

I learned Spanish using an MT course, and MT is extremely grammar orientated -- he uses hardly any fixed phrases. Because I had a good command of the grammar, I could just pick up the most common phrases and patterns from hearing and reading them.

In my experience, texts written in formal English are easily analysable into chunks, but in speaking, where native speakers express themselves in a less structured way and use the grammar that deviates starkly from textbook English, patterns are much more difficult to identify. So, the extent to which making a conscious effort to extract these patterns from the input you are exposed to helps you to improve may vary. I agree that regardless of your approach, in the long run you will end up thinking in chunks and not separate words just because this is how your brain processes language in general.

But you're agreeing with something I haven't said -- I have never said that this will happen regardless of approach, and I don't even believe that. Plenty of people fail to learn a language because the approach the teacher is using doesn't actually work. People who succeed often do it in spite of the teacher, not because of them.

The thing is, when you say "chunks" here, it's not even clear whether you're using the term as it was defined by psychologists, or just using it as a synonym of "fixed phrases".

See, the idea of fixed phrases has been around for a long time -- far longer than chunking. That's my real issue: it's another instance of the teaching profession glomming onto a new fancy idea... and then ignoring the actual meaning of it and just abusing it as a rebranding of something that's been done for aaaaaaaaaages.

Fixed phrases have been talked about for ages, and there's been a debate about how fixed they are.
In "I'll give him a piece of my mind", what exactly is the fixed phrase? Do we include the verb "give" or does the "him" that can be replaced with "her", "them", "you" etc stopped it being "fixed"? I think the suggested answer was calling them "semi-fixed phrases".

But then that also leads to the question of whether the fully-fixed phrases are lexicalised -- i.e. treated as a single word. Some are, but some aren't.

This is where the weakness and abuse of the term "chunking" kicks in -- as I say, chunks are built of chunks, and those chunks are in turn built of chunks, all the way down. When the word is used as a naïve euphemism for a fixed phrase, then all those layers are lost.
If a teacher builds a course around fixed phrases, they are not using chunking.

[Nikita]

But I'm not simply saying that teachers use the term differently from psychology -- I'm saying that teachers have picked up the term from psychology and keep saying that the psychology supports what they're doing, when it really doesn't.

You are absolutely right that chunking as defined by psychology isn't really something for the learner to worry about. My problem is phrase-based methodology doesn't teach chunking anyway, but chunking is used as a justification or rationalisation of the approach.

Now it is clear to me that you only disapprove misusing this term to create a false impression that there is scientific evidence confirming the effectiveness of the approach that doesn't really rely on chunking. The reason why I initially decided to take part in this discussion is because your description of chunking resembles how I see language learning in general.The idea that you should memorize phrases instead of words to become fluent has become very popular in recent years, yet people apply it differently to their learning process and often restrict it to memorizing colourful idioms that have little to no use in real-life conversations. Chunking, as far as I understand it, differs from that because it implies that our brain groups words together just to make them easier to remember and recall and it has nothing to do with whether they form an idiomatic expression or not.

The thing is, when you say "chunks" here, it's not even clear whether you're using the term as it was defined by psychologists, or just using it as a synonym of "fixed phrases".

I didn't use the term "chunk" to mean fixed phrases. It is actually the fact that you can modify them that makes the whole approach so effective. The main reason why I mentioned formal writing is because there you can find a lot of sentences that articulate a specific idea but can also be adapted to a different context. Informal speaking and writing, on the other hand, are full of expressions that are less flexible and thus more difficult to use.

See, the idea of fixed phrases has been around for a long time -- far longer than chunking. That's my real issue: it's another instance of the teaching profession glomming onto a new fancy idea... and then ignoring the actual meaning of it and just abusing it as a rebranding of something that's been done for aaaaaaaaaages.

Fixed phrases have been talked about for ages, and there's been a debate about how fixed they are.
In "I'll give him a piece of my mind", what exactly is the fixed phrase? Do we include the verb "give" or does the "him" that can be replaced with "her", "them", "you" etc stopped it being "fixed"? I think the suggested answer was calling them "semi-fixed phrases".

But then that also leads to the question of whether the fully-fixed phrases are lexicalised -- i.e. treated as a single word. Some are, but some aren't.

This is where the weakness and abuse of the term "chunking" kicks in -- as I say, chunks are built of chunks, and those chunks are in turn built of chunks, all the way down. When the word is used as a naïve euphemism for a fixed phrase, then all those layers are lost.
If a teacher builds a course around fixed phrases, they are not using chunking.

I think there is nothing wrong with teachers referring to phrases as chunks in language learning context, unless they do it to intentionally mislead their students. Basic language courses that suggest memorizing sentences from phrasebooks without understanding of how you can alter them when the need arises definitely should not use the word "chunking" in their description. But there are methods that are based both on studying grammar and learning common phrases, idioms and collocations that enable you to convey your thoughts more precisely. Strictly speaking, teachers who promote these methods also don't really use chunking. However, It seems to me that if they explicitly emphasized the distinction between "semi-fixed phrase" and "chunk", it would have no impact on the progress their students make.