There are a few small children, mostly girls, that learn to read very early. You read stories to them and before you know they’re reading by themselves. By very early, I men age 3 or 4.
Does this happen in China ?
I remember Jerry Pournelle mentioning on his blog years ago that he learned to read on his own before grade school.
I can recall learning to read, with my parents’ help, when I was about 4. But then I just stopped reading until I entered first grade at age 5, at which point I quickly picked up reading again and never stopped. I don’t know why I stopped reading. I can recall liking reading but I guess I was just more into other stuff.
English isn’t even optimized for this purpose. How quickly could children learn to read if we fixed English spelling to make some kind of sense. The Korean Alphabet was created for this specific purpose. My guess is that everyone learns to read much faster in Korea.
That has marginal utility because nobody reads by composing individual letters into words, often illustrated by the fact that a sentence with the letters in all words jumbled or with no vowels is not hard to decipher. You read by recognizing letter groupings. If we read by letter composition we’d never recognize misspelled words. It also doesn’t account for understanding homonyms.
(Where whole-word proponents go off the rails is thinking this can happen naturally if you teach beginners to memorize short words rather than applying English phonetic rules.)
The jumbled-letters thing isn’t true. Any example you’ve seen is contrived manually. I wrote code once to test it. With actual (pseudo-)random reordering of the letters, too few words were recognizable for the text to be read. The rule I followed is the the first and last letters stay the same, while the rest are reordered “randomly”.
Scrambling the letters makes the word visually unrecognizable.
Yes, the jumbled letter thing is true up to a point. The brain is very predictive in this sense with regard to syntax, lexis, and so on. When reading, we continually make rapid eye movements called saccades, typically lasting from 20 ms to 200 ms. If you are interested, eye-tracking research is explicative.The brain is never reading just what the eyes are on, anticipation typically moves to the next word or kanji or whatever is next, only slowing down for syntactically complex sentences in order to process.
I tested it. I tried it. The text was unreadable. That’s a fact.
“Buh buh gobba gobba but but but eye tracking gubba babba buh buh babble babble”
Who gives a shit? The text was unreadable. Unreadable. If your logic about something tangentially related “proves” otherwise, then your logic is bullshit, because the text was in fact not readable, in real life. Test the actual thing we’re discussing and call me back. Show me the code. Don’t just make up crazy hypotheses and pretend you’re talking about reality.
As lnog as you keep the frist and lsat ltteres and aivod spetarnetig the cultesrs of cosnnoants (lkie the dbloe “t” and the cnsooannt brfeoe the timenral “s”) the text is ullsauy unrdsetblnade.
(I admit it doesn’t work that well in english given how many permissible consonant clusters this language has)
This example weakened your argument. i had to double back and work at a fair bit of this.
Really? I parsed it just slightly slower that correct english….and english is not my native language. Everytime I see such examples it’s the same: not much trouble reading it fast, in fact it’s when i try to slow down to see individual words or groups of letter that I have more trouble. The trick seems to keep reading fast without getting slowed down on a particular word you don’t get at first…then this word is auto-completed from context and you keep going….
To be fair, as I wrote it I noticed it was not as easy to parse as the same example in my native language; hence my second comment. I guess this is because English has too many permitted consonant sequences (like, both “ght” and “gth” are possible arrangements) and many vowel arrangements that get different pronunciations (like “ee”, which reads different from “e”).
Thanks that was fun. Not too bad except spetarnetig which totally cratered my reading speed (and not due to what looks like an extra “t”). dbloe –> dbuloe, yes?
Correct on both counts. My bad.
Hand-optimized fake example, fuck off. I tested this for real. It’s bullshit. Go thou and do likewise.
I’ve thought about this a great deal and think you’re missing a few things.
There are a few strategies we use to read each language uses all the methods to a different extent, but the balance can be different. Strategy 1 is phonics, we associate symbols with sounds and build words. Some languages relay on this nearly 100% (Korean) others almost 0% (Chinese). After phonics we memorize exception cases of pronunciation, things like silent letters, or weird spellings from other languages that creep into yours. Another strategy we use is context, we are not sure what a word is but can figure it out based on our knowing the words around it, and clues about how it sounds from the other strategies. Finally, we use shape memorization to read quickly, this happens after we already know a language pretty well.
Now this is where whole language folks go off the rail, it’s not an appropriate method to teach children, you teach phonics, rules and context, and then kids get the shape memorization thing on their own. Once you figured out a word your brain chunks it into a shape and stores it. But it can only the chunk the information if the constituent parts make sense.
These things need to be taught in a particular order to work. You need to learn letters, then associated sounds, then the phonemes associated with syllables, finally you can learn the exceptions and use context. Your brain will do the chunking thing on it’s own as you figure out the words.
If you want to make a language easier to learn, you should reduce the number of shapes that need to be memorized, and the number of special cases that need to be memorized.
In english if we made capital and lowercase letters more similar, for the ones that are very different like Q & q, and used fewer letters and symbols that were just reflected versions of one another, think b & d. Kids could learn the alphabet more quickly. Additionally we can rid language of all silent letters, and start to reduce special cases. Kids would learn quicker
“These things need to be taught in a particular order to work. You need to learn letters, then associated sounds, then the phonemes associated with syllables, finally you can learn the exceptions and use context. Your brain will do the chunking thing on it’s own as you figure out the words.”
That seems like the best way to teach the overwhelming majority of kids how to read.
There must be some kids who learn how to read in spite of being subjected to the “whole language” idiocy, and maybe there are a very few kids for whom the “whole language” approach is actually better, but I just don’t see how “educators” pushing the “whole language” approach can be anything other than morons.
I read a paper once, an academic paper, about whole language learning. It was so transparently silly, I felt a deep sense of embarrassment for the authors. They were trying to critique the RCT evidence on phonics vs whole language by critiquing empiricism generally. Shame on whatever academic journal published that.
others almost 0% (Chinese).
Don’t be silly. You can very frequently guess the pronunciation if you know the pronunciation of one part, especially the non-radical part.
And Korean is not 100% phonetic. The spelling is about as weird as in English…and some letters sound different depending on whether they are first or last in a syllable. Plus, there are double letters and other letter combinations. And the letters look a little different depending on where in the syllable block they are.
But it’s a nice theory.
Can we quantize this? Might be that we’d get 5% Chinese vs 70% English
Hangul (written Korean) is very close to 100% phonetic. And the differences in how letters sound that depend on where they are in a sylllable is programmed into the letters’ names. If you learn the alphabet, you learn the pronounciation rules.
“Double letters” in Hangul are used to depict slight variations in phonemes. There are only five of them in an alphabet of 24 basic letters.
There are plenty of things happening in Korean phonology which Hangul orthography can only capture with as many convoluted rules as English. 사이시옷, for instance: explain to your L2 Korean learners why 나뭇잎 is [나문닙]. You can get there eventually, but it’s not that much easier than learning the rules behind the English bough, tough, and through. Then there’s ㄴ첨가: 옷입다 [온닙따], 못 잊어 [몬니저]. That 입다 also has 경음화, which is all over the place: 사건, 대가, 진가 [사껀, 대까, 진까]. That’s a separate phenomenon than 어두 경음화, which has been going on for 400 years in Korean and shows no signs of stopping: you may write it 주꾸미 or 수세미, but when 90% of people are pronouncing it 쭈꾸미 and 쑤세미, then the orthography is playing catch-up.
…and this is leaving behind stuff you can get away with: 의의 as [의이] OR [의의], different vowel length on 눈 ‘eye’ vs. 눈: ‘snow,’ etc…
Meihem in ce klasrum!
By the same logic Hanzi isn’t worse — we just recognize sets of strokes, insetad each one indiviually xD
Hangul (written Korean) is composed of letter groupings. The grouping is not fundamentally of words, but of syllables.
“That has marginal utility because nobody reads by composing individual letters into words”
research i read some years ago suggested some people see whole words and some just letters which if true and if good readers are mostly the former and therefore disproportionately go on to academia might explain why whole word teaching methods were a) accepted and b) such a disaster for most people.
English spelling sucks but still much better than Hanzi, a child can look word any unfamiliar word in dictionary fairly quickly. When I was a child dictionaries were most wanted books. And it doesn’t strain eyes so much.
Korea’s alphabet wasn’t in wide use before WW2, the Japanese wanted to assimilate Koreans. Until 19th century Koreans wrote with Hanzi
Reforming the English language to make it easier to read was a huge cause a century ago, with advocates such as George Bernard Shaw, Andrew Carnegie, and Col. McCormick, publisher of the Chicago Tribune. It’s interesting how most of these plans have died out, as has Calendar Reform. I was interested in both topics when I was a kid 50 years ago, but much of what I read back then was from old books from the library or that my mother brought home from thrift shop, so these reform movements might already have been dead by 50 years ago.
Reminds me of the attempts to get rid of the QWERTY keyboard, which also failed: too much investmente already in the old system.
I wonder. My experience in Japan was that most high school age kids could just barely read a newspaper, but I worked with average kids at best.
In Chinese schools they have dictionary lookup contests.
Slight off topic, but here’s a nice story. I know an eminent mathematician who is a third son. At some young age, he read something to his parents. Astonished, his parents asked how he learned to read. He told them son #2 taught him. Even more astonished, they asked how he learned to read. Son #1 taught him.
Granddaughters aged 5 and 3 read children books together, Not fluently, but Hebrew is more difficult than English. Chinese may be even more abstruse.
Good guess, but Terry’s the oldest.
Nothing about to be surprised here.
Now my father claims he learned to read when was stranded in a vegetable storage alone for a while. I think it’s false memory but it if it was true I’d been surprised.
I apparently learned to read really young (3). Both my parents were newly-minted special ed teachers, so they were probably practicing every technique they’d learned on me. My sister learned early but not quite as early as I did, and I remember helping to teach her.
My first two kids learned to read fairly early–both were reading before they got to kindergarten. My youngest didn’t learn to read very well until about 6, when she suddenly transitioned almost overnight from haltingly reading a word at a time to blazing through whole books. As far as I can tell, she never really got phonics and just memorized word shapes, and it took her awhile to get the needed vocabulary to get going. But also, she was very motivated to read all the Harry Potter books, so maybe she just finally found something interesting enough to make her care.
Precocious abilities suggest a trait is innate, so precocious literacy (in alphabetic writing systems) suggest that trait has been selected for. By the same token, if precocious logographic literacy exists, it must have been selected for as well, and the different cognitive architecture it relies on would go some ways in explaining the commonly observed psychometric differences between East Asians and Europeans.
I know this blog post wasn’t mean as a riddle, but have I read it right?
I think the idea is sounding out how quickly a logography can be learned, rather than underlying any psychometric differences (which are not implausible I guess but not really have any burning evidence for them).
Precocious children can learn to read quickly in English because they pick up verbal language quickly, and then can sideload that quickly into writing once the principle is demonstrated. In China its not obvious that translating general language skills is as easy.
That said I’d mostly guess that you do get spontaneous readers in China as well, more reliant on inference and dictionaries of course.
Do you mean does this happen with Chinese characters or does this happen with people of Chinese ancestry? I would assume that many Chinese kids know a few characters before they are taught to read. But of course you cannot generalise from a few Chinese characters to all the rest.
I learned to read long before kindergarten, which I entered a few months before I turned 5. Probably it wasn’t strictly “on my own,” but a kind of osmosis while people read aloud to me. Was there explanation or formal teaching? Unsure: my memory doesn’t go back as far as a period when the printed words made no sense. There are a few scattered memories of having to decipher something phonetically difficult like “ing,” asking for help, and being charmed by the answer: Oh! Those odd letters go with that familiar sound! I can still remember the shock when someone first suggested I could read silently to myself, which had never occurred to me. It was convenient, but also a bit sad.
Yes in hong kong.
“Children in Hong Kong are explicitly taught to write at a very young age. They are guided to draw vertical and horizontal lines at age three, and are required to write simple characters with few strokes at age four. When children are ready for formal schooling at age six, many of them can write around 50–60 characters. Typical writing lessons consist of penmanship drills and assigned character copying tasks”
Yes. This is about Chinese and the issue is writing. Not being a mathematician or a physicist but a bachelor of arts,I memorise Chinese characters as a hobby and I know that it is far, far more difficult to remember how to write (for instance) the place name Taiwan 臺灣 than to recognise it passively. The two mental abilities seem to involve different parts of the brain.
Anecdata but possibly of interest. My brother could not read until the age of nine. I have reason to believe, from his educational record, that he was in the 75th percentile of white British males.
Do the Chinese conduct spelling bees, and if so, how?
I (male) mostly taught myself through Sesame Street, during kindergarten. I could write my own name (in all-caps, and badly) aged six. I do remember that shoelaces were a problem until well into first grade.
I expect most regular-ish commenters here will be over the average in intelligence. I could well be the dunce here.
Writing is separate. Another whole group of skills is needed, including a certain amount of dexterity.
Once one can read, it is largely whole-word, enough so that we occasionally mentally mispronounce a word and have an “aha!” moment even years later when we hear it and realise it is a word we have known all along. I think the particular order of instruction idea is true, but some children are bright enough to race through steps almost instantly. I read brand names to my mother in the supermarket, likely using both letters and design from commercials (but not color, there was none on TV then) and leaping to what DUZ must be. Sometimes I would have to sound out longer words, but grasp it halfway through. I recall that for “tomatoes”
Accent can deceive. I was in fifth grade and read the word “parka” somewhere, and thought at first – “oh, that must be a winter garment sort of like a pahker,” and then instantly “Wait, are we saying that wrong around here?”
Yes. Logographic scripts are simply more difficult then alphabets or syllabarys
It was fashionable in some history circles, most prominently in Pomeranz’s ‘ the great divergence’ to emphasize how close China/Japan was to the west at the eve of the industrial revolution, particularly if one compared not the whole of China but various coastal provinces with Holland/England. One stat mentioned is literacy. However, this might overcount the actual literacy rate because of the phenomenon of fish literacy:
Fishermen might know all the relevants characters for fish-related profession, yet they wouldn’t be literate in our sense: able to tap into an enormous intellectual world, utilizing sources far displaced in space and time.
I have no idea when I learnt to read. I can’t remember not being able to read but then I have a lousy memory.
One of the few things I do remember from primary school is that when “Miss” taught us the two times table I went up to her and whispered that it went further: “two times sixteen is thirty-two, Miss”. Mind you, I don’t remember when I learnt to count or to “say my tables” either.
I wonder if reading ability esp at an early age coincides with these findings
Best part of this oftentimes oblivious article:
“the high school graduation rate among Japanese and Taiwanese students is around 95 percent, the report stated. But it’s below 70 percent for Cambodian and Laotian students….
The prevalence of the model minority stereotype distorts and renders invisible the experiences of these smaller groups,” the report’s authors wrote. “However, the few studies that are available on these subgroups reveal that they face significant structural barriers that often impede their educational achievement…and, thus, their life circumstances.”
Translation: We need more specific stereotyping to not mix up those Cambodians with the Chinese and Japanese. Genetics is a hell of a structural barrier after all…
I have the suspicion that the modern Latin script is easier to read than many others, including some other alphabets derived Greek exposure.
According to my theory, it influences literacy but is also sort of caused by literacy. The vast number of books written using the Latin alphabet influenced the evolution of modern typefaces or fonts. Some regions didn’t have many books printed and so their common font is more like handwriting and difficult to discern. Must be really difficult for people with poor eyesight.
In Latin, it was a while before there were spaces between words, which helps show how it evolved to be more readable. English definitely had some superfluous and difficult to read letters removed.
Cyrillic is hard FOR US because several consonants apply to Slavic phonemes which we don’t got. I assume that Russians can read Cyrillic just fine.
The Greek alphabet itself is less a problem that you’d think. After 2500 years Greek doesn’t feel as alien to the Western ear as does Slavic.
What I was really thinking of is more exotic scripts like Thai, Tibetan, and Mongolian. Maybe, I should have said “Phoenician” instead of “Greek.” I don’t know. We know the Greeks were in Asia, so I sort of like to attribute it to them instead. But, perhaps, Phoenician covers more bases, like Arabic, which seems hard to discern.
Part of it is no doubt recognition, but I think the letters are not as well-proportioned. One test is to go on Youtube with a smallscreen device and look at the comments in other scripts.
While I can read Cyrillic writing I find it hard. Of course that must in large part be due to lack of familiarity: I was in my late teens before I started trying to learn Russian and I still have read very little of it. But I think there’s another factor involved too, making the Cyrillic alphabet intrinsically harder to read than Latin-based alphabets: Too many Cyrillic characters consist of two or three vertical bars connected in various ways, and at a cursory glance they are hard to distinguish from each other. This also makes it difficult to read badly printed text, where the connections between bars may be invisible or nearly so.
I read cyrilic writing in my native language about twice as fast as English. Russian alphabet is almost the same as my country’s alphabet (give or take a few letters). But I read Russian very, very slowly. I use the common or international words and decipher the unknown by the context. So I think that your main issue with Cyrilic alphabets is that the reader is using his second(or third) language.
Anecdote— according to my parents I could fluently recite the Chinese characters in a newspaper at age 3 without knowing the meaning of any of it. That year I moved to another country having never learned the ability to write in Chinese. In my case there was a dissociation between sounds, meanings, and writing.
Interesting observation, and likely true. But the reason the latin typefaces are more readable is because moveable type printing was functionally invented and developed in Europe. Yes, someone might have used a similar process in China and Korea earlier, but their efforts were either not adopted with any continuity, or else were not used to effect mass communication but instead were treated as a curiosity for scholars or the wealthy few.
European type foundries , on the other hand, were developing type faces for every other known alphabetical language within 50 years of Gutenberg. in 1539, there was even a press in Mexico City printing in native American languages. Certain other regions either did not adopt printing or even actually outlawed it, thus confining printing in their languages to printers in places like Vienna, London, etc. The Wikipedia article about “Global Spread of the printing press” is very accurate at the moment, if not complete in detail.
Thus, the profession of typeface design often fell by default to people whose first language was written with the Latin alphabet. For example, this specimen shows some of the typefaces created by William Caslan in several languages in the 18th century (rather late in printing history): https://www.antiques-atlas.com/antique/a_specimen_of_printing_types_by_william_caslon/as167a828 Some alphabets have not yet experienced their own “golden age” of type design, but perhaps the internet is driving towards that with the unique needs of digital “type”. 🙂
I’m one. I can remember as far back as age 2 and can’t remember not knowing how to read.
Age 1 ½, baby
For earliest memory. I doubt I could read
C’mon man, you should be able to remember that shit.
I also learned to read at 3. Partly I learned from my parents, partly from educational video games and a Speak and Spell:
In China, schoolkids are first taught Pinyin, which is Chinese written with Latin letters. In Taiwan, they’re first taught Zhuyin, which is Chinese written with phonetic symbols graphically based on Chinese characters. And in Japan, they learn the Japanese syllabaries first. These phonetic systems are then used to teach the Chinese characters themselves. Books for kids are written in these phonetic letters or with the phonetic letters alongside the characters to show pronunciation.
I imagine some kids may pick up reading the phonetic letters by themselves. For the characters, they may pick up individual characters here and there, just like kids pick up numbers and symbols, but they probably don’t figure out reading whole sentences written in characters by themselves.
I learned to read before age 4. My parents read to me but weren’t really trying to teach me; at a relative’s house, apparently I was reading words on album covers to an aunt, and she asked my parents if they knew I could read. They did not. The next year, they bought a full set of World Book encyclopedias; I ended up reading all of them. In kindergarten, there was one other kid – a girl – who could read. Interestingly enough, twelve years later we ended up as the only National Merit finalists from our town, but at different high schools. She got the scholarship, I didn’t (I killed the ACT but the SAT was a different animal and the test prep industry didn’t exist yet in rural mid-south-west-central America; I did well but it didn’t measure up to my ACT score).
But in my small sample, I didn’t see much difference in reading abilities at the high end between girls and boys. At the other end, all of my grade school’s slow learners were boys.
ITT: lots of males confirm how their history proves girls can read at young ages
And nearly everyone forgets the original question was about China. 🙂
Larry Sanger (wikipedia cofounder) has a site and several blogposts on teaching his kids to read as toddlers: https://larrysanger.org/2010/12/baby-reading/
I was reading before I entered the first grade back in 1953. There was no kindergarten where I lived. I don’t remember learning, but I was read to (comic books!) primarily by my paternal grandfather and Little Golden Books by my mother. I was reading the newspaper about Jonas Salk and polio vaccine when I was in the second grade, and while I have no direct memory of that I was told about by my mother. I had an interest in polio as I had two classmates in a class of 35 who had been afflicted by it. Soon I was “fighting” my father for the sports page.
I’m certain I learned by recognizing words from having them read aloud to me and then extrapolating from there. I have long thought that the whole language approach must have been based on the observation that really good readers learned to read that way…. so that must be the best way to teach it. I regard that approach as the educational equivalent of the food pyramid foisted on us by our betters; it’s upside down.
One of my kids learned to read in the bathtub, I think starting before he turned two.
My wife had gotten foam letters that would stick to the walls of the bathtub, so while the kid was taking a bath, I’d spell out three-letter words. I don’t remember exactly how old he was when he could sound out any simple word, but I’m sure he could do it well before he turned three.
Do kids in Spain learn to read earlier, on average, than in places with harder spelling rules? Spanish is really nice for reading phonetically–there are a couple weird cases where you can get the same sound multiple ways, but in general the rules are pretty simple and if you see something written you know how to say it.
Spanish may be the easiest and the most logical language, thanks to centuries of simplifying effort by the Real Academia de la Lengua, By end of the first grade, everybody can read.
My two youngest daughters learned to read completely differently. The older learned the phonetic rules rather systematically and sounded things out. The other read English like Chinese–she memorized whole words. The two of them have different minds and characters. Evidence, of course, that genes are powerful. The are 16 months apart.
My earliest memory is of discovering that I knew how to read. We had new library books and I was pestering my mother to read them to me. She was washing the dinner dishes and told me to look at the pictures until she was done. I was looking at a two-page spread of two small children gazing into the window of a toy store and suddenly the black squiggles under the picture resolved into words. There was frost on the windows, so I think I had just turned four.
When I entered kindergarten the next fall, Miss Vail handed out workbooks with tear-out pages. Along the margin, on the stapled side, there were instructions (printed sideways in teensy type) and I happily followed them. Miss Vail was quite startled when I handed back my completed workbook a couple of weeks later.
When my son was in kindergarten, we took him to see the first Hobbit movie, and when we got home he asked whether we had that book.I said no, but we had a different hobbit book, and we started reading LOTR. When we finished Vol. 2, he told me not to start Vol. 3, because he’d already read it. He took it to first-grade show-and-tell as his first chapter book. This startled his teacher in turn. She had put him in the slowest reading group because he never knew where they were when it was his turn to read aloud. (Undiagnosed ADHD.)
I also started learning to read at 3. Of course, Spanish is relatively simpler. The most striking thing, however, is that I have a vivid recollection of the moment I learned to tell between my left and right hand: it was when I was seven years old, and it came as a flash.
Way off topic:
Any chance of Greg reviewing David Sloan Wilson’s new book? I’d pony up to help the cause. Listening to Razib Interviewing him now and going a little bonkers. Did he just say John Dewey was on the right track?! See you in Minneapolis in July.
I learn to read at 3. Mostly taught by my grandmother. I went on to earn a couple of academic degrees but I cannot say that my career has been remarkable, so in my case early reading is probably not related to intelectual brilliance.
All four of my kids learned to read by three or four, my second and fourth child primarily on their own. With the latter I didn’t even know he could read until he asked for a particular selection from a menu without prompting.
Around my 3rd birthday, I asked my mother to teach me how to read. She spent 2 weeks showing me how sounding out words worked and going through the children’s book “Are You My Mother?” with me, and turned me loose. The next book I read was “The Wizard of Oz”. My kindergarten report card listed my reading level as “Adult”.
Hyperlexia—The section on non-English studies has a bit on Cantonese and Korean studies:
Interesting it tends to be girls—it might be my imagination but my infant daughter and son when he was an infant both would seem to study my mouth and try to imitate the movements when I talked to them, and my daughter seems more prococious on this count.
My wife was an early reader but her brother was normal I think.
Fill in your details below or click an icon to log in:
You are commenting using your WordPress.com account.
( Log Out /
You are commenting using your Google account.
( Log Out /
You are commenting using your Twitter account.
( Log Out /
You are commenting using your Facebook account.
( Log Out /
Connecting to %s
Notify me of new comments via email.
Notify me of new posts via email.