What is more important?

[ablindwatchmaker]

@alantin 

 

I’ll quote when I figure out how to use it lol. 
 

It is much easier to get in reading and listening practice, for sure. Not only that, but trying to force output when you aren’t ready is extremely inefficient, as opposed to having the foundation in reading and listening and then speaking. 
 

@realmayo

 

Those are really interesting studies, further confirming what we’ve been saying. 

[Guest realmayo]

On 10/23/2021 at 11:51 AM, alantin said:

If word can be encoded in the brain as a sound unit

 

I suppose the thing to remember is that most people who have ever lived have never read. I believe that if you put people with no language on a desert island*, they will inevitably invent a spoken language or, if deaf, a sign language. But they probably won't spontaneously invent a written one. So it seems to make sense to me that the language we have evolved to use as communication between people is primary, while reading is secondary. Therefore it woudn't surprise me if the secondary activity builds on the primary one.

 

*or a new Nicaraguan school for deaf children: http://news.bbc.co.uk/1/hi/sci/tech/3662928.stm

[alantin]

On 10/23/2021 at 1:55 PM, ablindwatchmaker said:

I’ll quote when I figure out how to use it lol. 

 

You highlight the text you want to quote and it pops up a "quote selection" button at the end of the highlighted section.

[杰.克]

On 10/23/2021 at 11:51 AM, alantin said:

How about people who are deaf from birth

 

Yeah, I guess this question, unfortunately, is going to have to stick to fully able bodied people. Clearly there are going to be unique cases where the importance spectrum is upended. Noone is going to say listening is the most important, to a deaf person (thought that would be assumed already). 

 

We can still have an interesting conversation though!

[alantin]

On 10/23/2021 at 2:22 PM, 杰.克 said:

Yeah, I guess this question, unfortunately, is going to have to stick to fully able bodied people. Clearly there are going to be unique cases where the importance spectrum is upended. Noone is going to say listening is the most important, to a deaf person (thought that would be assumed already). 

My question was a comment to the point about words usually being coded in the human brain as sound units. However it seems to be possible to code them as visual information too. I'd be interested to see research on how this applies to Chinese people as I've heard many natives claim that they can just look at the Chinese characters on a page and understand what it means without converting them to sounds first. Also I know for a fact that many Chinese and Japanese people read their native languages at insane speeds. I can't even read my own native alphabet based language that fast ?.

 

So to me it seems that there are at-least three possibilities.

 

1. They are lying and actually translate the words to sounds like the rest of us, they just do it really fast,

2. they really do save time by skipping the sound part,

3. when they read fast, they are actually skimming like the rest of us.

[pinion]

On 10/23/2021 at 7:41 PM, alantin said:

I've heard many natives claim that they can just look at the Chinese characters on a page and understand what it means without converting them to sounds first.

There's also plenty of people who claim to read English without subvocalizing, so my conclusion is that self-reporting is unreliable, and that your 1) and 3) are true.

[Guest realmayo]

On 10/23/2021 at 12:41 PM, alantin said:

I've heard many natives claim that they can just look at the Chinese characters on a page and understand what it means without converting them to sounds first.

 

I think every language has people who will make that claim. [cross-posted with pinion]

 

On 10/23/2021 at 12:41 PM, alantin said:

words usually being coded in the human brain as sound units

 

I wasn't saying they were coded as sound units - I think that would imply the human brain works like a computer, which is utterly wrong (I believe).

[Guest realmayo]

On 10/23/2021 at 12:22 PM, 杰.克 said:

We can still have an interesting conversation though!

 

Here is perhaps a better version of your question:

 

Most people on these forums would expect a certain amount of progress over the next 24 months in reading/(writing)/speaking/listening. If I could inject you with 24 months of progress right now, but for only one of those four areas (and then you carry on studying and progressing as usual), which one of those four would you choose? Which is your booster shot of choice?

[Insectosaurus]

On 10/23/2021 at 12:37 PM, realmayo said:

Personally, I'm increasingly convinced that a negative effect of characters is that, if we learn new vocabulary from reading texts and dictionary definitions, we never 'force' that word into our heads as a sound unit - but just as a viusal code that we must decipher to discover its meaning. I think that's a problem. So these days when I'm learning new vocabulary, I'll make sure to test myself on pinyin->word and English->word as well as the more traditional 汉字->word.

 

This has been my experience as well. Now when I'm studying Japanese I have made sure to completely eliminate kanji from my active studying routine (to avoid the false comfort of using my knowledge of characters). This because listening and speaking competence will be much more important to me personally in the future when it comes to Japanese.

 

The problem with testing myself on pinyin is that Chinese has a huge amount of homophones. This can be solved in a number of ways, but I've so far not completely adored any of them. One is having the word in pinyin, then an example sentence (either in pinyin or characters), or having an entire sentence in pinyin. You can also add information about part of speech, but even that wouldn't cut it. Since I already have a passive (written) vocabulary probably exceeding 10,000 words, I don't see myself "restudying" a lot of words I don't hear, but will rather, as I do know, catch up by podcasts and music.

 

The optimal way would of course be native audio, which can be difficult if you encounter the word in a novel, a news article or other media without audio.

[Jan Finster]

Regarding listening vs reading: I believe both can be exhausting, but I can imagine active listening is more tiring, because while you can read at your own speed, you have to listen at a predetermined speed (unless you rewind or use 0.5-2x speed options). The great thing about listening is that you can do it passively. Yes, this is not the same as active listening, but you still learn something when you listen, e.g. while doing the dishes, driving or working out. You cannot really speak, read or write passively. 

 

Regarding writing: you can only write what you know actively. So, unless you are really advanced in reading and speaking, you cannot be advanced at writing (I cannot write like say, Christopher Hitchens, if I do not even know the vocabulary he uses...).

[alantin]

On 10/23/2021 at 5:00 PM, Jan Finster said:

Regarding writing: you can only write what you know actively. So, unless you are really advanced in reading and speaking, you cannot be advanced at writing (I cannot write like say, Christopher Hitchens, if I do not even know the vocabulary he uses...).

This is precisely why I believe neglecting writing to be a mistake. Input activities (Reading and Writing) generally broaden your passive vocabulary and a part of that passive vocabulary you know well enough that it turns into your active vocabulary, which you can use in output activities (Speaking and Writing). Actually finding ways to use new words in conversation will make those words slowly move from you passive vocabulary to active vocabulary. However because while speaking with someone, you'll have time constraints and generally can stop and search for a better word that you know exists but you just can't recall, I find that speaking will limit you to a much smaller active vocabulary than you have available while writing. Writing gives you time to think about how to express what you want to say more precisely and you can also search for words that you wouldn't be able to use while speaking. Writing creates a fast lane for those words to move to your active vocabulary and allows you to become familiar enough with a larger selection of words, than only speaking does, so that those words become available for you also while speaking. So I think you will always be able to write more than you would be able to say. Similarly, you will always be able to understand more input than you will be able to output yourself.

 

The direction of new words entering your vocabulary is in through input activities, of which reading is more effective, in regard to learning the language, because you can set the speed and take all the time in the world. Listening is more advanced but also a great input activity in this regard, but often you'll need check the word's written form too to be absolutely sure about it, so I regard reading as more basic or fundamental of the two for an adult language learner because of the "training wheels" it allows. Then finally the new words can enter the active vocabulary only from the passive vocabulary and like I described above, speaking is more advanced and demanding than writing.

 

This is how I generally view language learning but obviously with Chinese the hanzi create greater inertia with the written language than there would be with other languages. I think if someone said that they're learning German, but see no point in learning how to read and write it, most people would think they're nuts. In the case of Chinese (and Japanese), on one hand I believe it would be a lot more challenging to learn the language while limiting oneself to only one input and one output activity (Listening and speaking) but this is offset by having to put in the effort to learn a couple of thousand characters in the process if one wishes to also participate in reading and writing. How much this offset is is an interesting question. My own experience, having studied Japanese to fluency without focusing that much on reading and writing, is that you can end up very fluent but with a lot smaller vocabulary than you would have if you also read a lot. So I imagine in the short term only focusing on listening and speaking may seem beneficial, but in but in the long run it will cost you the benefits provided by reading and writing: Deeper understanding of the language, access to all the written native material out there, and the accumulating effect of passive and active vocabulary brought by extensive reading.

 

I might therefore argue that you might as-well get the characters over with in the beginning and at-least learn to read them if the effort is going to be about the same or only a little more (just differently distributed) relative to only focusing on listening and speaking. Also nowadays writing is a lot easier than it once was when you can just type pinyin and pick the correct characters, so essentially learning to read, gives you writing (or at-least typing) as an added bonus.
 

I regard my own not focusing on reading and writing in my Japanese studies as a mistake I don't want to make again and therefore I have focused extensively on the characters with Chinese. Of-course in my case I already knew about 1500 Japanese kanji when I began learning Chinese and I find the way Chinese uses the characters to be orders of magnitude easier than the way does, so I hardly began from zero in regards to reading and writing. Therefore my experience may not be comparable to someone with no prior knowledge of the characters and I don't know to what extent.

[phills]

On 10/23/2021 at 7:41 PM, alantin said:

So to me it seems that there are at-least three possibilities.

 

1. They are lying and actually translate the words to sounds like the rest of us, they just do it really fast,

2. they really do save time by skipping the sound part,

3. when they read fast, they are actually skimming like the rest of us.

 

It's a good question.

 

I don't know about others, but for me, since I started reading more, I started learning to associate some chunks of text directly from the characters -> meaning, without going through the sounds.  But that's only for relatively easy & familiar text.   I can do that for many parts of 兄弟, which is why I read that book near the top end of my speed range. 

 

(Another common example -- I do that for names of people in the book; I often just recognize them by the time I get near the end of the book and I no longer pronounce their names, which have no inherent meaning anyways.  So much so that when other people pronounce their names, it sounds weird to me, even though it's exactly the same way I would pronounce it; I just haven't for a while). 

 

However, whenever the text gets harder, or the patterns become unfamiliar, I slow down and start pronouncing every character (double checking it's correct).  Then I start trying to piece them together serially 1 at a time to understand it. 

 

So I think even if you never learned to listen, when the content gets complicated, if you're trying to piece together just visual characters, you'll downgrade to inputting the characters in your head one at a time, serially & in order.  You might never actually "speak" it (although sounding it out helps), but in essence, you're piecing it together as if it were audio. 

 

To me that's different mental process than solving a picture puzzle.  I don't think you could piece together meaning like a picture puzzle unless you had scratch paper.  If you had scratch paper, you have a lot more flexibility.  You could jump to the end of the sentence first, if you spot an easy word there, and then jump back to the middle, and translate that, and write down their equivalents on a scratchpad with ______ in between the translated bits for (unknowns).  The translation would become like a jigsaw, filling itself in as you figure out more stuff out.

 

But I don't think most people can juggle all that in their head.  They build context serially from the text in order, which is in effect a listening process even without sound.  I imagine people who are born deaf still have innate brain capacity for processing language this way.

 

When I have real problems with understanding a sentence, sometimes I do actually take out a pad and try to treat it like a picture puzzle, mapping the characters to words 1 at a time (using a dictionary as necessary) and disregarding order or ignoring clauses / subclauses.  Sometimes I cut and paste pieces in and out of google translate as I try to figure out which piece does what, which is similar to having a scratch pad.  That translation exercise is much clunkier, and I would view that as not equivalent to listening (although it still might be helpful for improving listening).

 

I suppose that's my theory of reading.  Maybe scholars who study ancient languages and never speak it, do treat reading like a jigsaw puzzle exercise; and because they've done so many translations on paper, they learned the language that way.  But it seems more natural and quicker doing it serially 1 character / sound / unit at a time, and humans already have dedicated brain structures to do that.  I think this applies most when you're dealing with text that you find hard but not impossibly hard; if it's easy text, you might grasp it all at once or out of order; if it's impossibly hard, you start diagramming and re-arranging. 

[Jan Finster]

On 10/23/2021 at 6:08 PM, alantin said:

So I think you will always be able to write more than you would be able to say.

 

I am not sure this is necessarily true. Basically, everything you write is formulated in your brain and in many cases you will actually subvocalize the words in your head (i.e. speak them) as you write the words down. The difference is you have more time for thought when you write.  If I told you to start writing 100 words per minute without the possibility of editing the words, I am not sure you would be able to express more than if I told you to speak at the same rate. On the other hand, if you are to give a speech and I tell you that you must not write anything down, but you can take all the time you want to formulate the sentences, then the results would quite likely be as good as if you wrote them. 

 

On 10/23/2021 at 6:08 PM, alantin said:

I think if someone said that they're learning German, but see no point in learning how to read and write it, most people would think they're nuts

 

With Chinese you have to differentiate between handwriting and writing on a computer. The former is of course the main discouragement for most learners. I totally believe in the benefits of writing, but not necessarily in the usefulness of being able to write by hand.

 

[alantin]

On 10/23/2021 at 9:50 PM, Jan Finster said:

I am not sure this is necessarily true. Basically, everything you write is formulated in your brain and in many cases you will actually subvocalize the words in your head (i.e. speak them) as you write the words down. The difference is you have more time for thought when you write.  If I told you to start writing 100 words per minute without the possibility of editing the words, I am not sure you would be able to express more than if I told you to speak at the same rate.

It think we agree on this. My point is that with writing you'll have more time to formulate what you want to say, which is usually not the case when you have someone sitting across from you expecting you to say something. How people usually speak also tends to be a grammatical mess if you read a transcript of someone talking normally or stop to pay attention to it, so if the speed was made constant 100 words per minute, I'd expect the spoken performance to far outweigh the written.

 

 

On 10/23/2021 at 9:50 PM, Jan Finster said:

With Chinese you have to differentiate between handwriting and writing on a computer. The former is of course the main discouragement for most learners. I totally believe in the benefits of writing, but not necessarily in the usefulness of being able to write by hand.

 

When someone says "writing" I generally understand it as "composing paragraphs of text either on computer or paper", but maybe for clarity this should be divided into typing and handwriting in these discussions.

 

 

@phills, very interesting to read about your experience. I'd expect the Chinese natives to do the same thing, but they'd just be able to process more and larger junks like this. However I still have to wonder; how does what you describe differ from skimming? At-least for me skimming is scanning paragraphs of text in non-linear fashion without subvocalizing or only subvocalizing only a couple of words per line. But if I want to really concentrate on the text and actually read it, I'll subvocalize pretty much everything.

[phills]

On 10/24/2021 at 3:32 AM, alantin said:

However I still have to wonder; how does what you describe differ from skimming?

 

I skim a lot in English, and the main difference to me is when I skim in English I actually don't read everything.  I'm somehow able to tell (not always correctly) if the words I'm skipping are valuable to my interest or not.  Somehow if I don't think they contribute to the plot, I just skip them.  E.g. I jump straight to their conversation and I pick out the sentence that says Person A hit Person B, but skip the part that said what color Person A's jacket was. 

 

So when I skim, I (mostly) grasped the plot, but if you asked me afterwards what color A's jacket was, I couldn't tell you.  I couldn't even tell you the author specified it.  On the other hand, when I read fast but avoiding skim, I could tell you.  Or at least I could tell you I forgot exactly but I remember the author mentioned it somewhere along the way.  When chunking, I do read almost everything, just in chunks.

 

A further example.  When I read novels with multiple storylines, I often skim the secondary storylines or the ones about the characters I'm not so interested in.  E.g. when I read Game of Thrones, I skimmed the Dany storyline for the few books, preferring the storyline in Westeros mainland.  When the Dany storyline got more interesting, I slowed down, and realized I missed a lot of details in the earlier Dany sections; but I still got the context that she was moving around freeing slaves in various cities with confusing names.  That's what I would term skimming as opposed to reading fast.

 

But I'm not a speed reader.  I never figured out how those people get to 700wpm+ in English.  I tried using their saccades and tried reading a whole paragraph at a time, but never achieved much success.  My comprehension level was too low for it to be worth it for me.  So maybe someone else can weigh in on those techniques.

[alantin]

@phills, thank you for clarifying!
This is really interesting stuff! So have I understood this right? You are saying that while reading in Chinese, you'll see chunks like 毫无疑问, or 虽然......但是 and just understand them without subvocalizing them but the same isn't true for English and you would subvocalize phrases like "therefore" or "without a doubt..." etc.

 

Would you say that this ability to read chunks of often recurring Chinese this way is something that has appeared spontaneously as a result of reading a lot or have you somehow worked for it?

 

If this is a naturally occurring phenomenon, it certainly is something to look forward to!

[Guest realmayo]

On 10/24/2021 at 10:36 AM, alantin said:

understand them without subvocalizing

 

I think it's a damaging myth that subvocalising is bad. Current thinking seems to be that you can't read properly unless you subvocalise. Wikipedia's entry on subvocalisation begins:

 

Quote

Subvocalization, or silent speech, is the internal speech typically made when reading; it provides the sound of the word as it is read. This is a natural process when reading, and it helps the mind to access meanings to comprehend and remember what is read, potentially reducing cognitive load.

 

If we're reading Chinese because we want to improve our Chinese, then it seems crazy to try to read in a way which makes it less likely that we'll process what we've read. Our brains can do a lot of magic if they're given comprehensible input, but the words on the page have to be properly inputted.

 

(That doesn't mean skimming is always wrong. First, it's an important skill. Second, if it allows you to skip the stuff you don't want to read, then you should be better able to engage with the good stuff that you do want to read once you get to it.)

[Guest realmayo]

On 10/23/2021 at 6:39 PM, phills said:

However, whenever the text gets harder, or the patterns become unfamiliar, I slow down and start pronouncing every character (double checking it's correct).  Then I start trying to piece them together serially 1 at a time to understand it. 

 

Just to be clear: as I understand it, yes, this is an example of subvocalisation, but quite an extreme one. You will also be subvocalising when you read in your native language at a quick, comfortable speed.

 

Perhaps (just me guessing) what slows us down reading Chinese is that we subvocalise too slowly, in part because we're not that familar with the characters. The goal must not be to eliminate subvocalisation, but to train ourselves to subvocalise so quickly that we don't normally realise we're doing it.

[alantin]

@realmayo, I'm in the camp that believes that you have to subvocalize in order to understand written language. Skimming on the other hand is reading some parts and guessing the rest.

 

However, I'm not quite sure yet if the same is true for Chinese. And after a very cursory examination, that Wikipedia article also doesn't seem to provide the answer. It seems to quote only non-asian sources so the question remains, is subvocalization as essential for readers of ideographic languages as it is for readers of non-ideographic ones?

To illustrate, do you need to subvocalize the ideograms in the picture to understand them?

 

Source: The Wikipedia page on "ideogram".
"Ideograms in the Church of the Visitation, Jerusalem"

[Jan Finster]

On 10/24/2021 at 10:59 AM, phills said:

I skim a lot in English, and the main difference to me is when I skim in English I actually don't read everything.  I'm somehow able to tell (not always correctly) if the words I'm skipping are valuable to my interest or not.  Somehow if I don't think they contribute to the plot, I just skip them.  E.g. I jump straight to their conversation and I pick out the sentence that says Person A hit Person B, but skip the part that said what color Person A's jacket was. 

 

So when I skim, I (mostly) grasped the plot, but if you asked me afterwards what color A's jacket was, I couldn't tell you.  I couldn't even tell you the author specified it.  On the other hand, when I read fast but avoiding skim, I could tell you.  Or at least I could tell you I forgot exactly but I remember the author mentioned it somewhere along the way.  When chunking, I do read almost everything, just in chunks.

 

A further example.  When I read novels with multiple storylines, I often skim the secondary storylines or the ones about the characters I'm not so interested in.  E.g. when I read Game of Thrones, I skimmed the Dany storyline for the few books, preferring the storyline in Westeros mainland.  When the Dany storyline got more interesting, I slowed down, and realized I missed a lot of details in the earlier Dany sections; but I still got the context that she was moving around freeing slaves in various cities with confusing names.  That's what I would term skimming as opposed to reading fast.

 

Yes, this is what I do when I read fiction. However, does it not defeat the purpose of fiction? To me good books of fiction should create a mental movie in your mind while you read them. This mental movie is created by the descriptions . If you just focus on context, you might as well just read a summary. For that reason I mainly read non-fiction, but I do watch  tons of fiction (Netflix, etc) 

 

On 10/24/2021 at 10:59 AM, phills said:

I never figured out how those people get to 700wpm+ in English.

 

I do not think they do. They are mostly scanning or skimming. My father was a college professor of literature and he could read a 200-300 page book on the level considered "literature" in less than 2 hours. I once asked him, how he can read so fast. He said he just skips all the descriptions of scenes.... 

 

On 10/24/2021 at 12:18 PM, realmayo said:

Current thinking seems to be that you can't read properly unless you subvocalise.

 

I do subvocalise a lot, but I I guess it depends on the text if it is necessary or not.

One of the books I am currently reading is The Naked and the Dead by Norman Mailer. The first couple of pages describe a scene of soldiers on the night before an attack. It is rich in descriptions about the stuffy heat, the fear, sleeplessness and of officers playing poker. I guess it is possible to read this without subvocalising and to have the scene appear as an image or a  movie in your mind.

The same cannot be said when I read abstract texts or non-fiction. Here subvocalising for me is a must and it helps me digest the content.