reading fluency

[semola]

I was wondering if there are researches on how many pages (or characters) a L2 student has to read to achieve reading fluency.

I remember a classmate of mine saying that if you read three big books (like 700 pages each) you should be able to achieve reading fluency in chinese.

Do anybody know anything about this?

(I've just finished my first big book, 明朝那些事, two to go! )

[Karin2HP]

明朝那些事 is a very good book but maybe a little hard for a beginner

[kdavid]

Your first question should be how to define reading fluency.

Second, I'm finishing up the first semester of my MA in Ancient Chinese History. All of the reading is in Chinese, and I've easily read over 2100 pages of Chinese text over the past few months. While I've seen a dramatic increase in reading speed and vocabulary, I'd hardly say I'm on par with a native in regards to reading fluency.

[skylee]

Your first question should be how to define reading fluency.

I agree.

明朝那些事兒 sounds like fun. I will take a look.

PS - there are many android apps on this book. if only I could find one in traditional Chinese ...

[roddy]

Measure your reading speed in characters per minute, and aim to increase that until you're at a level you're happy with. Imron's given figures of 300-700 cpm for native speakers, so you could see that as a goal to work towards.

[semola]

With reading fluency I meant being able to read pretty much anything smoothly and without using the dictionary much,so both speed and characters knowledge are implied.

I found 明朝那些事 really hard even though I don't consider myself a beginner, I would recomend easier and shorter books to start with, it's esay to get bored if you don't understand what you are reading for too long.

I found the pleco reader very useful for the reading purpose,especially on the Ipad.

@kdavid, impressive, are you doing your MA in Harbin?I'm finishing mine here in Kunming but I didn't have to read that much.加油

@roddy thanks for the answer, the one you gave is a good parameter to do some tests.

thanks,

Alberto

[imron]

With reading fluency I meant being able to read pretty much anything smoothly and without using the dictionary much,

Define "using the dictionary much". With a typical book, if you had 99% comprehension, that would still mean you'd need to look up 5-7 characters a page. Although you could definitely keep reading without looking them up, would that still be considered smooth reading?

Also, it depends a lot on the content of the book. Some books will use simpler language that you can breeze through easily. Others will use more formal or literary styles and they'll typically be harder to read (thus impacting smoothness of reading).

Personally, I found it took about 4-5 books (spanning approx 1 million characters) read in reasonably quick succession to start to feel like reading wasn't a chore, and about the same again to really start to feel comfortable reading novels - and this was from a base of already being quite comfortable reading newspaper articles and the like. That being said, despite having read 32 books in the last 2 years totalling some 9 million characters, my reading in Chinese is still far behind my reading in English in terms of comfort and fluidity. I'm still trying to figure out a good way to bridge that gap.

Suffice to say, I don't think 3 novels (even big ones) will cover it.

[skylee]

I think perhaps we don't need to get too academic about the definition of fluency. But I agree that the difficulty of the book will affect how fluently you can read it.

I have taken a look at the first few chapters of the book mentioned by OP. The language is very simple and easy. If the OP found that difficult then there is still a long way for him to go. But these things always take time and perseverance, and it was already a good start.

[Olle Linge]

Isn't "reading fluency" more a function of vocabulary and knowledge of the language in general, rather than the number of books you've read? Sure, fluency is dependent on prior experience and actual skill achieved only through reading, but vocabulary should still be all-important.

[skylee]

Besides knowledge on the language, knowledge on the country, its history, its literature, its culture, and the writers' styles also matters, I would think. And if you read genre novels like wuxia novels, or books on Chinese medicine or the Bible in Chinese, for example, your knowledge on these areas would also help you read more "fluently", I would think.

[imron]

Isn't "reading fluency" more a function of vocabulary and knowledge of the language in general

Yes and no. There's also another aspect, which is the process of actually taking the vocabulary and knowledge that you have and trying to apply it to whatever text you have in front of you to try and make sense of it in a way that is not disruptive to enjoying the reading process. Perhaps you could argue that this is covered by 'knowledge of the language', however I think it's not so much knowledge of the language, but rather just practical reading experience.

Assuming you're at a point where you can start to tackle native material, then regardless of your vocab level, I think there's always going to be an initial hump when trying to get into reading longer works, and it's only after you've started to accumulate reading experience that this goes away. You can see the phenomenon quite often, when people who have built up large vocabularies through various means, still have trouble if they try to read longer length native materials.

but vocabulary should still be all-important

Only up to a certain point, and it depends a lot on the material you are reading. Past a certain point, vocabulary differences are going to start to be more and more related to specialist fields/topics, so say one person has a vocab of 10,000 words and another person has a vocab of 12,000 words, but they are both reading a book that only contains vocab that they both know (not all that infeasible if they are just doing some easy reading, rather than reading on a specialist topic). In this case, I would argue that whoever has the most reading experience will find things more fluent, even if it was the person with the smaller vocabulary.

[semola]

Thanks for all the answers!

Ok, with reading fluency I meant something comparable to my english reading fluency (as you've certainly guessed english is L2 for me), I can read pretty much anything and I don't have to make an effort to understand what the writer is saying (this implies that looking up words doesn't affect my reading experience).

I agree with pretty much all the comments, I just find discomforting that after having read 32 books in 2 years imron doesn't consider himself fluent in reading, it's another proof that chinese is not easy.

@skylee I think what made me consider the book hard was not the register that the writer used but the words he used ( not commonly heard in the everyday language, like liutenant for ex) and especially my complete lack of knowledge on that period of chinese history.

To give another example I found that reading second language acquisition material in chinese was easier than reading mingchao, when almost all of my chinese classmates found our SLA textbook hard.I personally think that the reason why I found that book not super hard to read is because it was written by a foreigner and then translated in chinese,so the structure of the book felt somehow familiar to what I've read in the past.

Anyway, I think that the only way to be good at something is to do it over and over keeping an open mind, so I ll go back to my readings!

Hope people will keep posting their ideas.

[jkhsu]

I am a big proponent of using reading to acquire vocab because it helps me visualize the word or phrase better than a straight definition from a dictionary, even if it's a Chinese - Chinese dictionary. When I see a word or phrase, I try to remember how I saw it in another text to help me visualize it for the current text. Just the act of constantly creating different images of how the same word / phrase is used over and over in multiple contexts, I believe, helps one obtain a better grasp of the language.

[imron]

I just find discomforting that after having read 32 books in 2 years imron doesn't consider himself fluent in reading

The problem is more that I don't find myself as fluent at reading Chinese as I am at reading English (it probably took far more than 32 books to reach that level in English - my native language, so I'm not that worried yet). Basically I feel I can read anything written in modern standard mandarin without needing to resort to a dictionary and although there would still be words and characters I don't know, for the most part not looking them up would not impact my understanding or enjoyment of what I was reading (see the discussion here for some interesting data on this). I might still choose to look up those words and characters because otherwise I'll just be coasting by on my current level and not really improving, but I don't necessarily need to.

Missing vocabulary is not the only difference however between my Chinese and English reading levels. The other thing is effort. When I read English, there is zero effort required. When I see a word on the page, I don't need to spend any time processing it, I see it and instantly know its meaning without even needing to try. With Chinese however it feels like I actually need to spend time processing what I'm reading. Even though some words might be instant, for many other words there is still a tiny fraction of a delay (and for others perhaps an even longer delay), and for the most part there is almost always one or two of these words in any given sentence, and so that impacts the overall smoothness of my reading as it feels like constant stopping and starting and stopping and starting, without ever being able to get a nice long smooth run. What tends to happen then is that to avoid the stopping and starting feeling, I drop down my overall reading speed, but that is also not desirable in the long run, and it doesn't resolve the problem completely, it just minimises it.

Some people might suggest that it's unreasonable to expect my L2 reading skills to approach that of my L1, however the way I see it, my L2 reading has not yet reached a point that I am satisfied with and so I still see there is room for improvement. In recent years the general philosophy I've been following with learning Chinese has been:

1) Identify a weakness

2) Devise a drill to address that weakness

3) Drill that skill until it reaches an acceptable level

The link roddy posted above shows one of those drills, and while that has been great for my reading speed in general, I still find myself lacking that instant recognition I have with English. I've actually been thinking about this a lot recently, and think I might have found a way to address it, but want a spend a bit more time experimenting with it to see how well it works before discussing it in more detail.

[Lu]

After many, many books I still read faster in Dutch (L1) than in English (L2), but not by much, and I probably read English faster than many native speakers. So it's possible. But I've been wondering recently how I did it. Probably it was a matter of massive amounts of input and English being relatively easy to learn for a Dutch speaker. I've read some 10 Chinese books by now but am far from reading-fluent. I don't like looking up words so I rarely do, but I know I'm probably missing things because of that. So it's not just the number of books (and three is too few anyway), it's vocabulary plus practice. Large amounts of both, and you'll still read slower than in your native language, but you can read for pleasure and not miss anything (or hardly miss anything). Even more vocab and you can read to learn something new.

[rezaf]

I don’t know how many pages it takes to be able to read fluently and I am not there yet but I also think the biggest problem for most learners is weakness in vocabulary. An average Chinese university student knows around 40,000 words and even if we don’t include very professional words of one’s special field plus those that are not common at all, I think we still might have around 20,000 words that are commonly used in average books. Furthermore there are many different levels of knowing a word. Even if you know what it means, it still takes a lot of listening and reading until you can fully understand how it is used and more importantly how you can use it as part of your active vocabulary in writing and speaking. It’s only at that point that when you see that word in a book you can easily recognize it without needing to process it and even guess the coming words. Altogether I think it takes a huge amount of work to get to that level and for now I don’t want to worry myself about things like speed.

[Silent]

I think reading speed is ambiguous when it comes to reading fluency. The English language is more compact then the Dutch language. So in Dutch you've to read more to get the same information. For general reading I feel no difference in effort between English and Dutch reading. I never checked reading speed. Only for some more 'specialist' subjects I feel there is a difference. E.g. for me it's easier to read a grammar book in Dutch, but technical stuff is generally easier in English.

I think it's labourus to get (reading) fluency, but I see no reason why it would not be possible. It's mainly a matter of putting in the effort. As mentioned before you need to learn the vocabulary and a lot of practice. If you only start reading books after you've got your vocabulary at the right level, know your grammar, speak fluently and had some reading practice from reading bits and pieces I can imagine 3 decent books may be sufficient to get reading fluency as it's basicly only the reading longer text thing that has to be overcome. If you just pick up a book and start reading without any prior knowledge you won't manage in three books as those three books would have to be extraordinary only to cover the vocabulary needed for real reading fluency.

[imron]

I also think the biggest problem for most learners is weakness in vocabulary

This is true up to a certain point, but honestly I think for most material, the difference between a vocab of 20,000 and 40,000 will be negligible. Depending on the material, even the difference between 10,000 and 40,000 won't be so great, as long as you're getting over 98% percent comprehension (based on the figures and examples provided in the link above about the minimum levels of comprehension required to read something comfortably).

Furthermore there are many different levels of knowing a word.

I think this is an important thing. The level at which you know those words will play a key role in your reading fluency.

For example, do you recognise a word instantly without needing to do any processing, or is there a slight pause required for you to recognise and process that word? The unfortunate fact is, if you can't recognise it instantly without really requiring any additional processing, then it will affect your reading fluency to various degrees depending on how long it takes you to recognise and process that word.

The need for pausing versus instant recognition is also a determining factor of reading speed, and so although reading speed might seem ambiguous with regards to fluency, it does play a role, although actual speeds required will almost certainly differ based on the language. They key is whether you need to pause when you encounter a word, or whether you know it instantly.

Instantly is of course a poorly defined term as it will always require some time for your brain to process words, the trick is whether the brain does it fast enough that the processing is basically imperceptible to you. If we do the maths to try to see what sort of processing times are required, if you take 250cpm as a baseline for the bare minimum speed required for reading fluency in Chinese (approximately the average speaking speed of a native Chinese), you need to be spending less than 0.24s per character. To approach native level reading speeds, you probably need to spend a maximum of around 0.10-0.20s per character.

Unfortunately it's basically impossible to accurately test your recognition speed of individual characters to that degree of precision, however if when doing revision of vocab there is even a slight pause before recognising a word, then you have almost certainly exceeded those limits.

[Silent]

This is true up to a certain point, but honestly I think for most material, the difference between a vocab of 20,000 and 40,000 will be negligible. Depending on the material, even the difference between 10,000 and 40,000 won't be so great, as long as you're getting over 98% percent comprehension (based on the figures and examples provided in the link above about the minimum levels of comprehension required to read something comfortably).

It of course depends on the definition of 'word' and what you consider general reading material. I think about 3000 characters is considered basic literacy. In my search for suitable reading material I analyzed some books. I focused on material that I suspected to be fairly easy. In general I found books contain 2 to 3 times as many words as characters. So, based on that rule basic literacy is just a little shy of your 10000 words. I feel these figures match well with my results.

Around 3000 characters, 5k-10k words, seems quite normal for a book. With the above figures that seems an easy read. But this is only one book. Off course there is a fair overlap in vocabulary, but a fair difference too! E.g. I analysed 圈子圈套, now recommended as fairly easy. All three books combined came in at nearly 3400 characters and a massive 15000 words. The 98% comprehension is reached at well over 9000 words. Also considered fairly easy: 魔法禁书目录. I combined the first half (10) books of the series and got nearly 3500 characters and about 17500 words. The 98% comprehension is reached at again well over 9000 words. These are combinations of books in the same series, same genre. I didn't do the analysis (yet) but I think it's fair to expect that if these two, or other books from different writers and genre, are combined these figures go up again.

I feel a vocabulary of about 10000 words gives decent, but still fairly basic, literacy. I guess you will often be near the 98% comprehension rate and may be able to read fairly comfortable. To me however basic literacy is still shy of general reading fluency. With 10000 words however you may get fluency for the somewhat easier reading material.

[imron]

Basic literacy is considered 1500 characters for people living in rural areas and 2000 characters for people living in cities (source).

I would generally agree however that for reading Chinese, 10,000 words is probably a minimum vocab level required to be able to read general native material comfortably.

With my comment above though, I didn't mean to suggest that there will be no difference between someone who has a vocab of 10,000 vs someone who has a vocab of 40,000, just that at this level, for most reading material, I don't think vocab is going to be the most important factor for determining fluency.

I think your analysis of those texts, in conjunction with the demonstration in the video about 98% comprehension, would support this. With 10,000 words you would have 98% percent comprehension of those books and at that point you can more or less ignore the words you don't know without any significant impact on your reading. At that level, significant increases in your vocabulary aren't going to have a significant impact on comprehension or fluency (minor impact yes, but not major).