Kanji skimming
Nigel Greenwood
the use of kana & kanji for the benefit of English-speaking readers.
At one point he says that a quick scan of any Japanese document is
possible by skimming through it & reading only the kanji. "This too is
technology," he says (I'm quoting from memory).
Is this method of skimming widely used? If so, it's a useful tip for
foreign students of the language as well. (I can't help doing it in
any case, because I came to Japanese with a knowledge of Chinese.) In
practice, the use of kanji alone in newspaper headlines & titles is, I
suppose, the logical extension of this principle.
Nigel
--
ScriptMaster language resources (Chinese/Modern & Classical
Greek/IPA/Persian/Russian/Turkish):
http://www.elgin.free-online.co.uk
Wiktor S.
> foreign students of the language as well.
For foreign students who does not know many kanji, the above sentence may
read:
Is this something of somethinging somethingly used? If so, it's a useful
something for something somethings of the something as well.
For me, there's more information in kana and a small number of commonly used
kanji than in kanji alone.
--
Azarien
Phil Yff
> In his book Made in Japan, Sony's founder Akio Morita briefly explains
> the use of kana & kanji for the benefit of English-speaking readers.
> At one point he says that a quick scan of any Japanese document is
> possible by skimming through it & reading only the kanji. "This too is
> technology," he says (I'm quoting from memory).
>
> Is this method of skimming widely used? If so, it's a useful tip for
> foreign students of the language as well. (I can't help doing it in
> any case, because I came to Japanese with a knowledge of Chinese.) In
> practice, the use of kanji alone in newspaper headlines & titles is, I
> suppose, the logical extension of this principle.
I know that I find a document written in kana and kanji easier to read than
a document written in kana alone, even though the phonetics in a kana-only
document are easier. The eye is drawn to the kanji and because they embed
meaning, they quickly enable the reader to frame the fundamental issues of
the passage.
Phil Yff
Chris Kern
>Is this method of skimming widely used?
Possibly, but it's not very useful or successful. It's based on the
long-standing myth that Chinese characters represent ideas rather than
sounds and that somehow meaning can be extracted from them even if you
don't know the words in the language. Pure nonsense, but one that
shows up even in textbooks authored by language teachers.
-Chris
Nigel Greenwood
Perhaps I didn't make my meaning clear. I was wondering whether this
is a routine strategy adopted by native speakers/readers of Japanese
(to whom the kanji clearly do mean something!). Morita seems to imply
that it is.
As for the foreign students I referred to, I suppose I had in mind
reasonably advanced students with a good grasp of kanji.
Chris Kern
<ndsg...@yahoo.co.uk> posted the following:
>Perhaps I didn't make my meaning clear. I was wondering whether this
>is a routine strategy adopted by native speakers/readers of Japanese
>(to whom the kanji clearly do mean something!). Morita seems to imply
>that it is.
I don't believe so. Morita's statement is highly anecdotal, but I
don't know of any proof either way -- studies have shown that native
speakers rely much more on the sound values of the kanji than any
supposed "meaning" they might have, but that might not be relevant
here.
The problem with this statement is that I doubt it's possible for a
native speaker to scan over a document and literally read *only* the
kanji. For instance, if I gave you a few paragraphs were some of the
words were written in all capital letters and I told you to scan
through and only read the all-caps words and completely ignore all the
other words, I doubt you would be able to do that. What may be
happening here is that a Japanese native scans a document in the same
way that a native of any other language would do, and assumes that
they "only read the kanji" without that actually being the case.
-Chris
Oliver Cromm
> Chris Kern wrote:
>> On 20 Oct 2006 03:55:31 -0700, "Nigel Greenwood"
>> <ndsg...@yahoo.co.uk> posted the following:
>>
>>>Is this method of skimming widely used?
>>
>> Possibly, but it's not very useful or successful. It's based on the
>> long-standing myth that Chinese characters represent ideas rather than
>> sounds and that somehow meaning can be extracted from them even if you
>> don't know the words in the language.
It doesn't necessarily have to mean that. In the Japanese writing
system, there is a tendency that kanji represent word stems and kana,
grammatical information. Thus, reading only kanji can serve as a
grammatical filter, independently of the question of how kanji are
processed in the mind - I don't buy the direct-semantics theory myself.
In German, skimming text normally includes paying more attention to
capitalized words - I think that this is useful for a kind of
preliminary classification (but not an understanding - there's a danger
of abuse here), because there is more topical information in nouns than
in verbs and adjectives.
In Japanese, filtering for kanji is less selective than German
capitalization, but has a similar effect.
> Perhaps I didn't make my meaning clear. I was wondering whether this
> is a routine strategy adopted by native speakers/readers of Japanese
> (to whom the kanji clearly do mean something!). Morita seems to imply
> that it is.
It wouldn't surprise me if it was, but I have no hard data on that.
--
"Bother", said the Borg, as they assimilated Pooh.
Sean
> On 20 Oct 2006 12:50:11 -0700, "Nigel Greenwood"
> <ndsg...@yahoo.co.uk> posted the following:
>
>> Perhaps I didn't make my meaning clear. I was wondering whether this
>> is a routine strategy adopted by native speakers/readers of Japanese
>> (to whom the kanji clearly do mean something!). Morita seems to imply
>> that it is.
>
> I don't believe so. Morita's statement is highly anecdotal, but I
> don't know of any proof either way -- studies have shown that native
> speakers rely much more on the sound values of the kanji than any
> supposed "meaning" they might have, but that might not be relevant
> here.
Please provide citations for those studies.
>
> The problem with this statement is that I doubt it's possible for a
> native speaker to scan over a document and literally read *only* the
> kanji. For instance, if I gave you a few paragraphs were some of the
> words were written in all capital letters and I told you to scan
> through and only read the all-caps words and completely ignore all the
> other words, I doubt you would be able to do that. What may be
> happening here is that a Japanese native scans a document in the same
> way that a native of any other language would do, and assumes that
> they "only read the kanji" without that actually being the case.
>
> -Chris
--
Always be sincere, but never be serious.
Allan Watts
james...@gmail.com
ISTM that many Japanese do something akin to skimming, in that after
looking at a document for a few minutes they can give me a brief
description of the vague contents but not actually explain the details.
Real reading (in the sense that an English speaker would usually mean,
ie understanding to the level required to take appropriate action)
seems to require a lengthy forensic examination with a dictionary, or
better still enquiries of the originator of the document to find out
what he meant.
I found a short story in the toilet at work one day. It had furigana
over such words as "isu" (chair) that even I knew the reading for.
Mind you, I'm only surrounded by Todai graduates at a high-powered
scientiffic research institute in my day job, and it may be that the
hordes of the unwashed I see peering at the network maps at train
stations wondering out how the hell to get from A to B are perfectly
fluent :-)
I am beginning to suspect many Japanese go through life in a vague haze
of half-comprehension of the written world around them, and assume this
is normal for educated adults, and assume that without kanji it would
be even worse, because all they actually get is a vague impression of
the subject matter based on the kanji they recognise.
Asforhiraganaevenenglishisabitawkwardtoreadifyoutakeoutallthespacesdontyouthink?
Butwithpracticeimsureitwoudbecomemucheasier.
James
Chris Kern
following:
>
>Chris Kern wrote:
>> On 20 Oct 2006 03:55:31 -0700, "Nigel Greenwood"
>> <ndsg...@yahoo.co.uk> posted the following:
>>
>> >Is this method of skimming widely used?
>>
>> Possibly, but it's not very useful or successful. It's based on the
>> long-standing myth that Chinese characters represent ideas rather than
>> sounds and that somehow meaning can be extracted from them even if you
>> don't know the words in the language. Pure nonsense, but one that
>> shows up even in textbooks authored by language teachers.
>
>ISTM that many Japanese do something akin to skimming, in that after
>looking at a document for a few minutes they can give me a brief
>description of the vague contents but not actually explain the details.
Oh, I'm definitely not denying that. What I'm skeptical of is that
native Japanese readers skim by looking only at kanji, and that this
is some inherent property of the writing system that allows them to do
such skimming. Many literate people can skim documents no matter what
their native language.
>I am beginning to suspect many Japanese go through life in a vague haze
>of half-comprehension of the written world around them, and assume this
>is normal for educated adults, and assume that without kanji it would
>be even worse, because all they actually get is a vague impression of
>the subject matter based on the kanji they recognise.
I have heard this sort of viewpoint expressed by serious academics, so
you're not alone in thinking this. I'm not sure I think it's this bad
-- while I have no idea how literacy in Japan *on average* would
compare against some other country, I have a feeling that if you
compare an American college graduate with a Japanese college graduate
you would find similar levels of literacy, for better or for worse.
-Chris
Chris Kern
following:
>On 2006-10-20 13:08:48 -0700, Chris Kern <chris...@gmail.com> said:
>
>> On 20 Oct 2006 12:50:11 -0700, "Nigel Greenwood"
>> <ndsg...@yahoo.co.uk> posted the following:
>>
>>> Perhaps I didn't make my meaning clear. I was wondering whether this
>>> is a routine strategy adopted by native speakers/readers of Japanese
>>> (to whom the kanji clearly do mean something!). Morita seems to imply
>>> that it is.
>>
>> I don't believe so. Morita's statement is highly anecdotal, but I
>> don't know of any proof either way -- studies have shown that native
>> speakers rely much more on the sound values of the kanji than any
>> supposed "meaning" they might have, but that might not be relevant
>> here.
>
>Please provide citations for those studies.
Geez, don't make me do that. :) I won't be at school until Monday but
if I remember I'll try to get the specific reference then.
The study involved giving native speakers a reading passage that
contained two types of errors -- one type in which a word was written
with an incorrect kanji that had the same pronunciation (i.e. 訂出
instead of 提出), and another type in which a word was written with an
incorrect kanji that had the same meaning, but a different
pronunciation (i.e. 良意 instead of 善意). They used machines to
measure the eye movements of the people to see when they "got stuck"
on a passage and had to go back, and found that they got stuck a lot
more on the second type of error than the first. In other words,
native speakers seem to not even notice as long as the sound value of
the kanji is correct.
I'm not sure why the results of such studies would be surprising,
given that language is biologically sound-related and that writing
systems develop to record the sounds. There are innumerable studies
and lots of evidence to support the conclusion that we "subvocalize"
when reading (not literally moving lips, but that the muscles involved
in speech are moving and that the speech area of the brain is
necessary for comprehension of reading), such as this one:
http://eric.ed.gov/ERICWebPortal/Home.portal?_nfpb=true&_pageLabel=RecordDetails&ERICExtSearch_SearchValue_0=EJ481168&ERICExtSearch_SearchType_0=eric_accno&objectId=0900000b8003a2b7
-Chris
Kevin Wayne Williams
> I am beginning to suspect many Japanese go through life in a vague haze
> of half-comprehension of the written world around them, and assume this
> is normal for educated adults, and assume that without kanji it would
> be even worse, because all they actually get is a vague impression of
> the subject matter based on the kanji they recognise.
I go through my day-to-day life in a vague haze of half-comprehension of
the written world around me, because it is generally written in three
languages that I don't fully comprehend. It is surprising how well full
comprehension can be faked.
KWW
Jeremy
I think I may have just stumbled across the study you are mentioning, or
one very close to it (Funny enough, I stumbled across it while trying to
find a full text copy of the article you mentioned, which I couldn't :(.)
But the study you mention above is:
"Writing errors in Japanese kanji: A study with Japanese students and
foreign learners of Japanese"
By T. Hatta, A. Kawakami and K. Tamaoka. Reading and Writing, Vol10.3-5,
Oct1998. ISSN: 0922-4777 (Print) 1573-0905 (Online)
Abstract:
The present study examined kanji errors in handwriting made by Japanese
students and Australian learners of Japanese. First, a cognitive
psychological model to explain the production of writing errors was
proposed based upon the analysis of 374 writing errors of two-morpheme
(kanji) compound words generated by Japanese students in spontaneous
sentence writing situations. Despite the common assumption that kanji
writing errors may not be related to the sounds of kanji characters
(i.e., morphological phonology), the present study found that
phonologically-related kanji writing errors were most numerous (60.0%),
followed by orthographically-related errors (43.6%) and
semantically-related errors (29.7%), including some overlap of these
three types. Second, 408 kanji writing errors made by students learning
Japanese in an Australian university were analyzed. Unlike the Japanese
students, these subjects wrote more non-existing kanji and made
orthographically-related mistakes rather than semantically- and
phonologically-related errors. This result must be related to the level
of kanji writing skills held by learners of Japanese. In light of these
results, several suggestions were proposed for the methods of teaching
kanji writing.
Hope thats what you were referring to.
-Jeremy
Chris Kern
following:
>I think I may have just stumbled across the study you are mentioning, or
>one very close to it (Funny enough, I stumbled across it while trying to
>find a full text copy of the article you mentioned, which I couldn't :(.)
No, the one I was talking about involved reading done by native
speakers, not writing. Like I said, when I go into school on Monday
I'll ask the teachers who told me about it for specifics on the study
and hopefully I can get the exact cite.
However, that does involve some similar conclusions. In some sense
this is a difficult issue because it's hard to say what the difference
is between a character communicating a meaning directly, and
communicating a meaning through association with sound in the brain.
-Chris
Low Zhen Lin
> In his book Made in Japan, Sony's founder Akio Morita briefly explains
> the use of kana & kanji for the benefit of English-speaking readers.
> At one point he says that a quick scan of any Japanese document is
> possible by skimming through it & reading only the kanji. "This too is
> technology," he says (I'm quoting from memory).
>
> Is this method of skimming widely used?
I would like to draw an analogy with Braille. There are a few grades of
Braille, with each higher grade having more abbreviations and
contractions. People who are used to a particular grade of Braille find
it slower to read a lower grade because they are not used to reading the
words spelt out in full.
I personally find it difficult and slow to read fully kana texts,
because I am used to seeing certain words in kanji - my mind cannot
automatically make the connection between what I see and what it means.
Ben Bullock
"Low Zhen Lin" <lo...@hotmail.com> wrote in message
news:45398...@news.tm.net.my...
> I would like to draw an analogy with Braille. There are a few grades of
> Braille, with each higher grade having more abbreviations and
> contractions. People who are used to a particular grade of Braille find
> it slower to read a lower grade because they are not used to reading the
> words spelt out in full.
>
> I personally find it difficult and slow to read fully kana texts,
> because I am used to seeing certain words in kanji - my mind cannot
> automatically make the connection between what I see and what it means.
First of all, the same question I asked Phil, what fully kana texts have you
read? I've never seen any fully kana texts except for children's books, so
I'm surprised that you've read enough of them to come to such a conclusion.
Second, what do you do when you have to listen to Japanese? Do you find it
difficult and slow to understand spoken Japanese, which doesn't have any
kanjis?
--
sci.lang.japan FAQ: http://www.sljfaq.org/afaq/afaq.html
Kevin Wayne Williams
>
> First of all, the same question I asked Phil, what fully kana texts have
> you read? I've never seen any fully kana texts except for children's
> books, so I'm surprised that you've read enough of them to come to such
> a conclusion.
>
There are a number of texts for JSL students that are written in this
fashion. I used to take the exercises from JFBP and rewrite them in
kanji so that I could read them more easily.
> Second, what do you do when you have to listen to Japanese? Do you find
> it difficult and slow to understand spoken Japanese, which doesn't have
> any kanjis?
>
Completely different thing. I recognize the *sound* of someone saying
(for example) すいようび because I am used to hearing it. I recognize
the sight of 水曜日 (or, in context, just 水) because I am used to
seeing it. When I see the string すいようび I have to stop, sound it
out, hear what I say, and go "Aha! 水曜日!", and proceed. I find reading
katakana dense text miserably slow going for equivalent reasons.
I believe that there are readers that are strictly hukt on foniks, and
they don't see that the fuss is about. For people that sightread a
language, presenting a familiar word in an unfamiliar orthography makes
them read it in a completely different manner.
Most of this discussion is old territory. Just google up Jed Rothwell
and you can entertain yourself for days.
KWW
Ben Finney
> the long-standing myth that Chinese characters represent ideas
> rather than sounds and that somehow meaning can be extracted from
> them even if you don't know the words in the language. Pure
> nonsense [...]
I'm not sure why you say it's nonsense. There are a great many words
written with kanji that I can read the meaning of, but have no idea
how to pronounce (because I haven't learned the pronunciation yet).
Moreover, I know not a jot of any Chinese pronunciation, but my
knowledge of Japanese kanji allows me to understand the meaning of a
wide variety of words written with Chinese characters. So, in that
sense, the characters are evidently representing meaning directly to
me, and not sounds.
If, on the other hand, you are saying that the symbol-to-meaning
mapping is one that didn't *naturally arise* with Chinese characters
without going through a meaning-to-sound mapping first, I can
certainly entertain that notion. It's just not what you seem to be
saying.
--
\ "An idea isn't responsible for the people who believe in it." |
`\ -- Donald Robert Perry Marquis |
_o__) |
Ben Finney
Tak To
> On Fri, 20 Oct 2006 22:56:25 GMT, Sean <se...@fake.com> posted the
> following:
>
>>On 2006-10-20 13:08:48 -0700, Chris Kern <chris...@gmail.com> said:
>>
>>>On 20 Oct 2006 12:50:11 -0700, "Nigel Greenwood"
>>><ndsg...@yahoo.co.uk> posted the following:
>>>
>>>
>>>>Perhaps I didn't make my meaning clear. I was wondering whether this
>>>>is a routine strategy adopted by native speakers/readers of Japanese
>>>>(to whom the kanji clearly do mean something!). Morita seems to imply
>>>>that it is.
>>>
>>>I don't believe so. Morita's statement is highly anecdotal, but I
>>>don't know of any proof either way -- studies have shown that native
>>>speakers rely much more on the sound values of the kanji than any
>>>supposed "meaning" they might have, but that might not be relevant
>>>here.
What does "rely on the meaning" mean?
>>Please provide citations for those studies.
>
> Geez, don't make me do that. :) I won't be at school until Monday but
> if I remember I'll try to get the specific reference then.
>
> The study involved giving native speakers a reading passage that
> contained two types of errors -- one type in which a word was written
> with an incorrect kanji that had the same pronunciation (i.e. 訂出
> instead of 提出), and another type in which a word was written with an
> incorrect kanji that had the same meaning, but a different
> pronunciation (i.e. 良意 instead of 善意).
What does an "error" mean in the latter case?? One cannot help
but wonder if the study is comparing apples and oranges.
> They used machines to
> measure the eye movements of the people to see when they "got stuck"
> on a passage and had to go back, and found that they got stuck a lot
> more on the second type of error than the first. In other words,
> native speakers seem to not even notice as long as the sound value of
> the kanji is correct.
>
> I'm not sure why the results of such studies would be surprising,
> given that language is biologically sound-related
Except for sign languages of course. Perhaps it would be more
accurate to say that except for the deaf, we all acquire our L1
via audio.
> and that writing
> systems develop to record the sounds.
Writing system of natural languages -- but that would make the
tautology obvious. OTOH, it is not clear if writing systems of
mathematical equations were developed to record sound.
Moreover, it is not clear if in reading, after the initial
stage of visual->mental audio->meaning, most of us do not
also convert to visual->meaning in favorable conditions.
(A related question is if there are different degrees of
"mental audio" activities in the brain.)
> There are innumerable studies
> and lots of evidence to support the conclusion that we "subvocalize"
> when reading (not literally moving lips, but that the muscles involved
> in speech are moving and that the speech area of the brain is
> necessary for comprehension of reading), such as this one:
> http://eric.ed.gov/ERICWebPortal/Home.portal?_nfpb=true&_pageLabel=RecordDetails&ERICExtSearch_SearchValue_0=EJ481168&ERICExtSearch_SearchType_0=eric_accno&objectId=0900000b8003a2b7
Tak
--
----------------------------------------------------------------+-----
Tak To ta...@alum.mit.eduxx
--------------------------------------------------------------------^^
[taode takto ~{LU5B~}] NB: trim the xx to get my real email addr
Jeremy
The error in the second occurrence is in different pronunciation but
same meaning. To further explain his example, 善意 is correct, and EDICT
says it means "good faith." A construction of 善=Virtue and 意=idea. 良
意 is not a word, but when you look at the literal meaning of the kanji
it portrays similar meaning -- 良=Good and 良=Idea.
I hope thats what you were confused about.
Paul D
While I don't answer for Mr. Low, I'd point out that all literate
adults read written language according to word shapes, and not
phonetically. Reading phonetically from character to character is a
tiring process that the brain abandons as soon as possible.
We can assume that once a Japanese speaker (or learner) gets accustomed
to the kanji shape of words, it takes longer to recognize the kana
shape. Naturally, you can still sound it out phonetically and get your
answer, but that probably takes longer yet. I'm sure nearly every
person reading this newsgroup can attest that reading kanji is easier
once you've learned them.
Written language often (and always, some argue) represents sounds; but
immersive reading is almost never a sound-based process.
Paul
Nigel Greenwood
Jeremy wrote:
> But the study you mention above is:
> "Writing errors in Japanese kanji: A study with Japanese students and
> foreign learners of Japanese"
> By T. Hatta, A. Kawakami and K. Tamaoka. Reading and Writing, Vol10.3-5,
> Oct1998. ISSN: 0922-4777 (Print) 1573-0905 (Online)
>
> Abstract:
Don't bother with the abstract -- just forward the Japanese text so
that we can skim the kanji ...
Ben Finney
> Written language often (and always, some argue) represents sounds;
> but immersive reading is almost never a sound-based process.
Thanks. That seems right. But, to be properly scientific, what can you
point us to as evidence to support this assertion?
--
\ "Beware of and eschew pompous prolixity." -- Charles A. |
`\ Beardsley |
_o__) |
Ben Finney
Chris Kern
following:
>Tak To wrote:
>> What does an "error" mean in the latter case?? One cannot help
>> but wonder if the study is comparing apples and oranges.
>
>The error in the second occurrence is in different pronunciation but
>same meaning. To further explain his example, 善意 is correct, and EDICT
>says it means "good faith." A construction of 善=Virtue and 意=idea. 良
>意 is not a word, but when you look at the literal meaning of the kanji
>it portrays similar meaning -- 良=Good and 良=Idea.
Right. If the kanji really communicated meaning *instead of* sound,
we would not expect native speakers to be bothered by such errors.
Both kanji compounds should communicate the same idea, regardless of
the fact that "ryoui" is not a word.
-Chris
Tak To
Right, but who has ever claimed that kanji conveys no sound
information?
The study said only that readers' eye movement "got stuck", which I
take to mean they hesitated or paused at the unfamiliar combination
of kanjis. There is no mentioning of whether the readers eventually
understood the intended meaning. In all likelihood they treated it
as a neologism, and this does not invalidate the theory that kanji
communicated meaning (as well as sound).
In contrast, a reader who has spot a homophone would treat it
as an error rather than a neologism.
Ben Bullock
news:2006102120085016807-paul@hiddenfortressten...
> Written language often (and always, some argue) represents sounds; but
> immersive reading is almost never a sound-based process.
Chris Kern seems to think differently, did you read his posts?
For me the easiest thing to read is romaji Japanese. I'm a little suspicious
of the motives of foreign learners who claim that they find kanji easier to
read than romaji or kana.
Paul Blay
> Chris Kern seems to think differently, did you read his posts?
>
> For me the easiest thing to read is romaji Japanese.
Good grief man! What the heck sort of screwball language acquisition
process did you go through?
> I'm a little suspicious of the motives of foreign learners who claim that
> they find kanji easier to read than romaji or kana.
It's been a long while since I last read kana only (_with_spaces_) so I can't
really comment on that but romaji* is not only difficult to read it gives me
literal headache. IIRC (AIMN) my speed** at reading Japanese novels is
approximately half that of my speed in English (by the page). I think that's
reasonable for a non-native. I'm not going to try to measure my romaji
reading speed because (a) I don't have any text long enough in romaji
and (b) I'm not going to torture myself just to prove a point.
* If you want to be picky, "Japnese in romaji".
** Assuming I don't stop to look up every word I don't know (typically one
or two per page).
Ben Finney
> "Paul D" <pa...@hiddenfortress.ten> wrote in message
> news:2006102120085016807-paul@hiddenfortressten...
> > Written language often (and always, some argue) represents sounds;
> > but immersive reading is almost never a sound-based process.
>
> Chris Kern seems to think differently, did you read his posts?
>
> For me the easiest thing to read is romaji Japanese. I'm a little
> suspicious of the motives of foreign learners who claim that they
> find kanji easier to read than romaji or kana.
Having learned the kanji, I definitely find the kanji version of a
text easier to read than without. The easiest explanation I can come
up with is the huge number of homophones in Japanese, which are
directly disambiguated by the kanji meanings.
In spoken language, of course, the meanings of homophones are
indirectly disambiguated by context (accents, situational, non-spoken
communication, etc.)
As Chris Kern would no doubt agree, our brains are evolved to weave
all this context to disambiguate spoken language and extract
meaning. In written language, though, much of the context of speaking
is lost, and in a language like Japanese with its high level of
indistinguishable-by-orthography homophones, the meaning carried by
the kanji allows easy disambiguation, allowing the reading to flow
much better.
There's also the point already raised in this thread, that
kanji-with-okurigana gives a visual structure to a written passage
that is lacking in a pure-kana passage. In a romaji passage, spaces
are normally used, and my eyes are already well-trained to find visual
structures in words consisting of the roman letters.
Thus, pure-romaji (with appropriate spaces) would perhaps be easier
for me to read than pure-kana (with spaces in the same places); but
both of them are appallingly lacking in necessary information --
meaning *and otherwise* -- compared to a kanji-with-okurigana passage.
--
\ "Broken promises don't upset me. I just think, why did they |
`\ believe me?" -- Jack Handey |
_o__) |
Ben Finney
Chris Kern
posted the following:
>The study said only that readers' eye movement "got stuck", which I
>take to mean they hesitated or paused at the unfamiliar combination
>of kanjis.
Yes, but they did this significantly more often with the
meaning-switch mistakes than they did with the sound-switch mistakes.
>In all likelihood they treated it
>as a neologism, and this does not invalidate the theory that kanji
>communicated meaning (as well as sound).
If they "treated it as a neologism" and it had nothing to do with the
sound/meaning issue, we would expect equal trouble with either type of
mistake. Kanji definitely "communicate meaning", the question is
whether they somehow communicate meaning independent of sound. This
experiment would seem to suggest that for fluent native speakers, the
sound value of the characters is much more important to reading than
the meanings.
-Chris
j...@csse.monash.edu.au
night. My wife recently bought some opera DVDs in Manila for about
$1 each (piracy, shmiracy), so last night we watched Ambroise Thomas's
"Mignon". My French aural comprehension is not great, and the DVD was
only subtitled in Chinese(Mandarin) and Japanese, so I turned on the
Japanese ones. It made all the difference; in fact on occasions my
(French-speaking) wife asked what was just sung. Often the subtitles
whipped fast far to quickly for me to read them entirely, so I found I
was zooming in on the kanji and kanji jukugo to get the key elements
before working on the kana and grammar. Of course it was a bit
repetitive: いとしいミニョン .... 愛している ... etc. etc...
--
Jim Breen http://www.csse.monash.edu.au/~jwb/
Clayton School of Information Technology,
Monash University, VIC 3800, Australia
ジム・ブリーン@モナシュ大学
Ben Bullock
"Ben Finney" <bignose+h...@benfinney.id.au> wrote in message
news:87bqo5w...@benfinney.id.au...
> Thus, pure-romaji (with appropriate spaces) would perhaps be easier
> for me to read than pure-kana (with spaces in the same places); but
> both of them are appallingly lacking in necessary information --
> meaning *and otherwise* -- compared to a kanji-with-okurigana passage.
You're the third person now to come up with such a statement. The problem
for me is that I don't understand how you, Low Zhen Lin, and Phil ever
manage to understand spoken Japanese. In a book I once read by a famous
journalist about "how to write understandable Japanese" (a book for native
speakers), one of the pieces of advice which really stuck in my mind is that
he said "never write Japanese which cannot be understood when read out
loud".
Chris Kern
<bignose+h...@benfinney.id.au> posted the following:
>Having learned the kanji, I definitely find the kanji version of a
>text easier to read than without. The easiest explanation I can come
>up with is the huge number of homophones in Japanese, which are
>directly disambiguated by the kanji meanings.
It's just because you're used to reading the kanji-kana mix. The
homophones thing is really a non-issue; there are very few honomyms
that cannot be disambiguated due to context.
>In spoken language, of course, the meanings of homophones are
>indirectly disambiguated by context (accents, situational, non-spoken
>communication, etc.)
I'm curious why you think that context does not exist in written
language as well. Are you honestly saying that if a native Japanese
person reads some written material out loud without making any
gestures, another native speaker will be unable to understand it
because of homophones?
>In written language, though, much of the context of speaking
>is lost, and in a language like Japanese with its high level of
>indistinguishable-by-orthography homophones, the meaning carried by
>the kanji allows easy disambiguation, allowing the reading to flow
>much better.
Can you give some examples of *actual* sentences using these
homophones that cannot be disambiguated via context (i.e. not
sentences purposely constructed like this). If there are really so
many homophones, it should be very easy for you to go to the Asahi
Shinbun site and pick 5 examples from the articles in just a few
minutes.
-Chris
j...@csse.monash.edu.au
>"Ben Finney" <bignose+h...@benfinney.id.au> wrote in message
>news:87bqo5w...@benfinney.id.au...
>> Thus, pure-romaji (with appropriate spaces) would perhaps be easier
>> for me to read than pure-kana (with spaces in the same places); but
>> both of them are appallingly lacking in necessary information --
>> meaning *and otherwise* -- compared to a kanji-with-okurigana passage.
Well, I find "pure-romaji" with or without spaces more difficult than
"pure-kana". Single words in romaji are ok, but whole sentences don't
seem like real Japanese.
>You're the third person now to come up with such a statement. The problem
>for me is that I don't understand how you, Low Zhen Lin, and Phil ever
>manage to understand spoken Japanese. In a book I once read by a famous
>journalist about "how to write understandable Japanese" (a book for native
>speakers), one of the pieces of advice which really stuck in my mind is that
>he said "never write Japanese which cannot be understood when read out
>loud".
When I have had to give presentations in Japanese I usually prepare
my speaking notes with far more kana that I would for a written text,
because I can't afford to be stumped for a pronunciation. Often they end
up with kana-words with the kanji pencilled in above so I can see at
a glance what it actually means. Sort-of furigana in reverse.
Kevin Wayne Williams
>
> "Ben Finney" <bignose+h...@benfinney.id.au> wrote in message
> news:87bqo5w...@benfinney.id.au...
>
>> Thus, pure-romaji (with appropriate spaces) would perhaps be easier
>> for me to read than pure-kana (with spaces in the same places); but
>> both of them are appallingly lacking in necessary information --
>> meaning *and otherwise* -- compared to a kanji-with-okurigana passage.
>
>
> You're the third person now to come up with such a statement. The
> problem for me is that I don't understand how you, Low Zhen Lin, and
> Phil ever manage to understand spoken Japanese.
Don't leap to conclusions. I prefer kanji/kana over kana or romaji, but
that doesn't mean my verbal skills are anything to write home about. My
reading comprehension has always been better than my aural
comprehension, precisely because the written version gives me better
grammar clues than verbal Japanese.
KWW
Chris Kern
<benkasmi...@gmail.com> posted the following:
>You're the third person now to come up with such a statement. The problem
>for me is that I don't understand how you, Low Zhen Lin, and Phil ever
>manage to understand spoken Japanese.
When I was studying on my own, I was focusing almost exclusively on
learning kanji, vocabulary, and building my reading proficiency. I
was satisfied if I understood the meaning of the sentences and did not
focus all that much on whether I was getting the readings of the kanji
correct. I did reach a point where I could read some things, but I
would not be able to understand them if spoken and could not read them
fluently aloud. I'm pretty sure that the explanation for this is that
without intending to, I was connecting these kanji sequences primarily
with English words rather than Japanese words, and when I read a
Japanese text I was sort of switching back and forth (unconciously)
between English and Japanese as I was reading. However, once I lived
in Japan and got a lot more proficient in the spoken language, my
reading ability and speed improved quite a bit as well, and I found
that I could understand romaji texts (such as the innumerable romaji
e-mails I received from colleagues) much better than I could before.
I would expect that someone not used to reading romaji or kana texts
(with spaces -- I don't know where these mythical "all kana" texts
people keep talking about come from) would read them more slowly than
a normal text just because they are not used to it, no matter what
their level of proficiency. But I think that if you actually have
trouble comprehending (even at slow speed) a Japanese text written in
romaji or kana that you would be able to read fine in normal
orthography, it probably indicates a weakness in general Japanese
ability rather than properties of the writing systems involved.
It's the same case as if I wrote English with an odd mix of capital
letters and lower-case letters, and with varying numbers of spaces
between the words. As a native speaker, I think that while your
reading *speed* would be affected by the unusual orthography, it would
not cause any problems of actual comprehension of the meaning of what
was written.
Maybe this isn't true, but it does have support from linguistics.
During the period I was talking about in the first paragraph, I was
certain that my ability in reading kanji-kana texts was because of
some inherent properties of kanji that made them easier to deal with
than romaji or kana. But I no longer think that's the case.
-Chris
j...@csse.monash.edu.au
>... I don't know where these mythical "all kana" texts
>people keep talking about come from ....
Bookshops. That's where I bought "こいぬのくんくん" by "ディック．
ブルーナ". They're for pre-schoolers.
Ben Bullock
<j...@csse.monash.edu.au> wrote in message
news:OIC_g.51010$rP1....@news-server.bigpond.net.au...
> Apud Chris Kern <chris...@gmail.com> (sci.lang.japan) hoc legimus:
>>... I don't know where these mythical "all kana" texts
>>people keep talking about come from ....
>
> Bookshops. That's where I bought "こいぬのくんくん" by "ディック．
> ブルーナ". They're for pre-schoolers.
I think Chris probably knows that pre-school books and books for first-year
elementary schoolers are written in kana using spaces (wakachi-kaki) rather
than using kanji and furigana, but it doesn't really explain why people here
have mentioned how hard it is to read books in kana, unless it really is so
hard for them to read these pre-school books which are all-kana.
Incidentally I recently (partly) read a book from the library:
and the author made exactly this complaint about not liking reading
children's books written all-in-kana.
Low Zhen Lin
> First of all, the same question I asked Phil, what fully kana texts have
> you read? I've never seen any fully kana texts except for children's
> books, so I'm surprised that you've read enough of them to come to such
> a conclusion.
Pokémon, or as it is called in Japan, ポケットモンスター. The games are
written in 100% kana with spacing and I find it impossible to skim-read
as I do with the English versions of the games. On the other hand if I
were to do close reading it would probably take just as long as with
close reading with kanji.
> Second, what do you do when you have to listen to Japanese? Do you find
> it difficult and slow to understand spoken Japanese, which doesn't have
> any kanjis?
To some extent, that is true - my grasp of spoken Japanese is
substantially weaker than written.
Louise Bremner
> I don't know where these mythical "all kana" texts
> people keep talking about come from
The initial stages in the testing here:
<http://www.amazon.co.jp/gp/product/B000GUFX06/sr=1-1/qid=1161509757/ref
=sr_1_1/503-3105852-2509542?ie=UTF8&s=videogames>
Without spaces, too.
________________________________________________________________________
Louise Bremner (log at gol dot com)
If you want a reply by e-mail, don't write to my Yahoo address!
Ben Finney
> "Ben Finney" <bignose+h...@benfinney.id.au> wrote:
> > Thus, pure-romaji (with appropriate spaces) would perhaps be
> > easier for me to read than pure-kana (with spaces in the same
> > places); but both of them are appallingly lacking in necessary
> > information -- meaning *and otherwise* -- compared to a
> > kanji-with-okurigana passage.
>
> You're the third person now to come up with such a statement. The
> problem for me is that I don't understand how you, Low Zhen Lin, and
> Phil ever manage to understand spoken Japanese.
As I said in the above message, in spoken conversation there are many
more channels of communication occurring than can be conveyed in
phonetic orthography: particularly pacing, stress, tone and so
on. These all serve to disambiguate the possible different meanings
carried in a stream of syllables.
--
\ "What's another word for Thesaurus?" -- Steven Wright |
`\ |
_o__) |
Ben Finney
Ben Finney
> Apud Ben Bullock <benkasmi...@gmail.com> (sci.lang.japan) hoc legimus:
> >"Ben Finney" <bignose+h...@benfinney.id.au> wrote:
> >> Thus, pure-romaji (with appropriate spaces) would perhaps be
> >> easier for me to read than pure-kana (with spaces in the same
> >> places); but both of them are appallingly lacking in necessary
> >> information -- meaning *and otherwise* -- compared to a
> >> kanji-with-okurigana passage.
>
> Well, I find "pure-romaji" with or without spaces more difficult
> than "pure-kana". Single words in romaji are ok, but whole sentences
> don't seem like real Japanese.
Yes, it would be a close call. I don't read many passages in pure
kana, so haven't really got any empirical (or anecdotal) evidence
there.
--
\ "War is God's way of teaching geography to Americans." -- |
`\ Ambrose Bierce |
_o__) |
Ben Finney
Nigel Greenwood
illuminating for me as a beginner in this challenging language.
I wrote:
> In his book Made in Japan, Sony's founder Akio Morita briefly explains
> the use of kana & kanji for the benefit of English-speaking readers.
> At one point he says that a quick scan of any Japanese document is
> possible by skimming through it & reading only the kanji. "This too is
> technology," he says (I'm quoting from memory).
>
> Is this method of skimming widely used?
While several of the contributions are clearly from professionals with
excellent Japanese, we haven't -- as far as I can see -- had many
replies from native speakers. It would be interesting to hear their
reactions to Morita's assertion.
The skimming method works for me at present: in fact I can't help doing
it, since with my background in Chinese the kanji cry out to be read
amongst the (still unfamiliar) kana. But no doubt this will change
with time.
On a different point, what exactly is the objection (apart from saving
paper) to using gaps to separate words in kana? Granted, it's not
traditional -- but then ancient Greek, & to some extent Latin, used to
be written without gaps, yet modern Greeks & Italians are happy enough
to use them!
Paul D
<bignose+h...@benfinney.id.au> said:
> Paul D <pa...@hiddenfortress.ten> writes:
>
>> Written language often (and always, some argue) represents sounds;
>> but immersive reading is almost never a sound-based process.
>
> Thanks. That seems right. But, to be properly scientific, what can you
> point us to as evidence to support this assertion?
If you really insist, I'll find you some academic papers. :)
However, it's well known — particularly among the typesetting industry
— from analyzing people's reading behaviour that your eyes don't look
at every letter when you read immersively. In fact, they don't even
look at every word! Your eyes skip along, with the fovea (centre of
vision) focusing on 5-7 spots per line of text, and the parafovea
taking in the shapes of the surrounding words at once, which your brain
interprets as words you know.
In fact, I have heard that the French elementary school system teaches
words whole, by shape, instead of using phonics, for this reason.
Paul
The Wanderer
> I don't know where these mythical "all kana" texts people keep
> talking about come from
In addition to the places other people have mentioned to date: the
scripts of at least some earlier video games, particularly RPGs, are
written entirely in kana. (The first one which I know of which used
kanji in text rather than in graphics was Final Fantasy V.) They do
generally have spaces at vaguely appropriate junctures, and tend to be
divided into comparatively small blocks anyway by the unusually low
limit on the number of characters which can be fit onto a line, but that
does not make reading them a trivial exercise. I've made more or less
partial attempts at several different games by now, and the only thing
which might take more of the time I've spent on that effort than trying
to parse the text (once recognized) into something I can understand is
disambiguation - figuring out which "words", i.e. which kanji, a given
block of kana correspond to.
--
The Wanderer
Warning: Simply because I argue an issue does not mean I agree with any
side of it.
Secrecy is the beginning of tyranny.
Chris Kern
following:
>Apud Chris Kern <chris...@gmail.com> (sci.lang.japan) hoc legimus:
>>... I don't know where these mythical "all kana" texts
>>people keep talking about come from ....
>
>Bookshops. That's where I bought "こいぬのくんくん" by "ディック．
>ブルーナ". They're for pre-schoolers.
Well, what I meant was that people are talking like they've been
confronted with all kana texts numerous times and been unable to read
them. The most contact I have had with all kana texts are old Famicom
and early Super Famicom video games.
-Chris
Chris Kern
<bignose+h...@benfinney.id.au> posted the following:
>"Ben Bullock" <benkasmi...@gmail.com> writes:
Context is powerful enough that it disambiguates things in writing
too. As I said in the other post -- if there are really so many
homophones in Japanese, you should be able to go to Asahi's news site
and very quickly come up with 5 sentences that would be
incomprehensible to native speaker when read aloud (in an
expressionless voice with no gestures) due to the homophones in the
sentence. (Assuming the sentences are read in the context of their
articles.)
-Chris
Paul Blay
> Well, what I meant was that people are talking like they've been
> confronted with all kana texts numerous times and been unable to read
> them.
I get confronted with _romaji_ relatively often, generally from Japanese
learners wanting advice on translation.
> The most contact I have had with all kana texts are old Famicom
> and early Super Famicom video games.
<aol>Me too</aol>
Kevin Wayne Williams
What textbooks did you people use? JFPB doesn't use any kanji at all
until the end of book 2, and never gets to the point where the text
looks at all normal.
KWW
Chris Kern
posted the following:
>On 2006-10-21 20:22:02 +0900, Ben Finney
><bignose+h...@benfinney.id.au> said:
>
>> Paul D <pa...@hiddenfortress.ten> writes:
>>
>>> Written language often (and always, some argue) represents sounds;
>>> but immersive reading is almost never a sound-based process.
>>
>> Thanks. That seems right. But, to be properly scientific, what can you
>> point us to as evidence to support this assertion?
>
>If you really insist, I'll find you some academic papers. :)
>
>However, it's well known ? particularly among the typesetting industry
>? from analyzing people's reading behaviour that your eyes don't look
>at every letter when you read immersively. In fact, they don't even
>look at every word! Your eyes skip along, with the fovea (centre of
>vision) focusing on 5-7 spots per line of text, and the parafovea
>taking in the shapes of the surrounding words at once, which your brain
>interprets as words you know.
Right, but this does not mean it's not still a sound-based process.
As I said in another post, when a linguist says that reading is a
"sound-based process" he doesn't mean that you literally move your
lips or hear a voice in your head, or that your reading speed is
limited by the speed at which you can hear something. Obviously none
of that is true. All it means is that the process of reading relies
on the language ability that you have acquired through auditory means,
and that as such, it's not really correct to speak of "bypassing
sound" to somehow get directly to meaning.
When you write "your brain inteprets as words you know" -- there is no
such thing as "words you know" aside from your primarily audio-based
comprehension of language. Even if you initially learn a word through
the eye, in a text, your brain is still going to attempt to assign a
pronunciation to it of some sort, even if it's not the correct one.
There is a lot of proof for this, but one of the most evident is the
way that prelingual deaf children cannot be taught to read as easily
as even postlingual deaf children. If there really was some "meaning
area" of the brain that was totally independent of the spoken language
and that could be accessed directly through written symbols, we would
not expect prelingual deaf children to have any more trouble learning
to read than anybody else.
-Chris
Chris Kern
<ndsg...@yahoo.co.uk> posted the following:
>While several of the contributions are clearly from professionals with
>excellent Japanese, we haven't -- as far as I can see -- had many
>replies from native speakers. It would be interesting to hear their
>reactions to Morita's assertion.
The problem with that is that native speakers wouldn't necessarily
have anything conclusive to say. As a native speaker of English, I
cannot describe in detail the method by which I skim something (i.e.
which words I tend to look at, how much comprehension I have, etc.) I
wouldn't expect native speakers with no linguistic background to be
able to do that either.
>On a different point, what exactly is the objection (apart from saving
>paper) to using gaps to separate words in kana? Granted, it's not
>traditional -- but then ancient Greek, & to some extent Latin, used to
>be written without gaps, yet modern Greeks & Italians are happy enough
>to use them!
There is no objection. To the extent that all-kana text ever appears
in native Japanese materials, it always appears with spaces (unless
you are dealing with a particularly low-memory electronic device -- I
remember my fax machine in Japan that had messages in all katakana
with no spaces.)
-Chris
Chris Kern
<kww.n...@verizon.nut> posted the following:
Started with JSL/JWL. Their philosophy is not to introduce any text
that is not orthographically authentic. Everything in JWL is written
like a native speaker would write it (obviously the linguistic content
is simple, but in terms of orthography, only what can be represented
authentically is introduced).
The text in JSL is all in romaji, but the textbook is more like a
reference manual for the CD-ROM, which is audio-based. The romaji in
JSL is intended primarily for reference, and learning to read romaji
is not a goal of the book.
-Chris
Low Zhen Lin
> On a different point, what exactly is the objection (apart from saving
> paper) to using gaps to separate words in kana? Granted, it's not
> traditional -- but then ancient Greek, & to some extent Latin, used to
> be written without gaps, yet modern Greeks & Italians are happy enough
> to use them!
Methinks the issue is not so much using spaces as it is where to put the
spaces. One camp will put spaces after particles, but not before, the
other camp will put spaces around particles.
Mike Wright
That's great advice, but how many Japanese writers actually follow it?
I agree that *if* Japanese were written to be understood when read
aloud, there is no reason that a fluent speaker shouldn't be able to
understand it perfectly well when reading a romanized version of it (or
kana with word spacing)--assuming at least as much early training and
practice as is required by speakers of Italian and Spanish. Punctuation
should normally be able to provide enough sentence rhythm information.
Of course, the written language will always have a little more potential
for ambiguity than the spoken language, but it should seldom be a huge
problem. The problems are practical, involving training and history, not
technical/theoretical.
I feel the same way about the Chinese languages. Some progress was made
in the writing of Christian materials in romanized Hokkien. I have a
character dictionary that has all the "definitions" in romanization, and
it was aimed at native speakers.
For those of us who are less-than-fluent non-native speakers, kanji can
be very helpful. Like Nigel, I learned Mandarin first, and was familiar
with about 2,000 characters before I started learning Japanese. I often
understand written Japanese that I can't even pronounce. (I'm sure it's
been at least eight years since I've heard a word of Japanese spoken,
and about 27 years since I've heard it with any regularity--or done much
reading.)
Of course, this is not an approach I'd advise anyone to take on purpose,
but the reality is that kanji helps break kana up into manageable
strings. My experience with learning to read Arabic tells me that the
reading of kana with word spacing could be learned fairly quickly.
As far as Nigel's original point goes, I do *not* find that I get much
out of skimming text by reading only the kanji. For one thing, most of
the syntax is embedded in the kana. Sure, you might get some idea of the
subject matter, but things like tense and negation are essential to
actually understanding a written passage.
Also, as Chris said, I doubt that a native speaker could "literally read
*only* the kanji". A native speaker would surely be able to imply more
from context than I can. Seeing 対 (tai) alone in a sea of kana, a
native speaker might not have to focus on the surrounding kana in order
to guess that what is written there is に対して (ni taishite). However,
I wonder if they could really resist looking at the surrounding kana for
context. This is really the kind of question that is answered better by
experiment than by intuition.
I just tried some mixed text in a variety of books (biography, art,
technical), and found that how much I could get by concentrating on
kanji varied with the writing style. Some works are full of long
technical terms written in kanji, and I could get a good idea of the
subject matter, though I didn't get much of what was being said *about*
the subject. In others, the kanji is pretty sparse, and I couldn't get
anything out of those.
It's probably easier for me to totally ignore the kana than it would be
for a native speaker, since it's been so long since I've spent any
significant time reading Japanese.
--
Mike Wright
http://www.raccoonbend.com
Mike Wright
I read pretty fast in English, but I still sort of hear the words in my
head. In fact, if I don't get that feeling, I know that I'm not really
taking in what I'm reading. (To give you an idea of how much I read, the
summer after I graduated from high school, I averaged a bit over 3.5
books--mostly novels--a day for a period of about 75 days.)
The more the text involves complex new facts, the more I seem to need
the connection to the sound of it in order for it to really sink in.
> When you write "your brain inteprets as words you know" -- there is no
> such thing as "words you know" aside from your primarily audio-based
> comprehension of language. Even if you initially learn a word through
> the eye, in a text, your brain is still going to attempt to assign a
> pronunciation to it of some sort, even if it's not the correct one.
I agree.
> There is a lot of proof for this, but one of the most evident is the
> way that prelingual deaf children cannot be taught to read as easily
> as even postlingual deaf children. If there really was some "meaning
> area" of the brain that was totally independent of the spoken language
> and that could be accessed directly through written symbols, we would
> not expect prelingual deaf children to have any more trouble learning
> to read than anybody else.
And the reason that most non-native students have trouble with long
strings of kana is that they are not really trained in reading it. The
goals of many courses include speaking and kanji recognition, but it
seems that few actually focus on reading kana text. They think that all
that is necessary is to be able to recognize the individual kana. But
the reality is that this leaves you reading one kana at a time, not
seeing them as whole words. I suspect that native speakers get beyond
this due to the amount of material that they have to read, but many
foreign students won't read as much in a lifetime as a Japanese student
reads in a year. (No need to jump forward with all the exceptions.)
Chris Kern
posted the following:
>And the reason that most non-native students have trouble with long
>strings of kana is that they are not really trained in reading it. The
>goals of many courses include speaking and kanji recognition, but it
>seems that few actually focus on reading kana text. They think that all
>that is necessary is to be able to recognize the individual kana. But
>the reality is that this leaves you reading one kana at a time, not
>seeing them as whole words.
This is often one of the great failings of Japanese language programs
-- "reading" is equated with learning characters, and so "mastering
the kana" becomes simply a factor of whether you can associate the
kana with the correct roman letters, and "learning a kanji" means
being able to write it, know the readings, and know the "meaning" that
the textbook has assigned to it. Whether you can actually read the
characters in context is only of secondary importance.
-Chris
Sean
>
> Yes, it would be a close call. I don't read many passages in pure
> kana, so haven't really got any empirical (or anecdotal) evidence
> there.
I imagine there's a few us in here (Ben is one I bet) who have had a
lot of experience reading kana-only children's books aloud to our kids
when they were very little. It's really tough at first, but with
practice it gradually becomes more fluid. The spacing is interesting.
Spaces were between phrases. Sometimes it was just word + particle,
sometimes longer phrases. Anyway, reading kana-only text is a skill
like any other. Practice gets you there.
Teaching Japanese to high school kids, I definitely get them reading
aloud for a portion of every class. After a while you can tell that
they're beginning to read in phrase chunks, not kana by kana or word by
word.
The kids invariably say that they find it easier to read passages that
have a few kanji in them, just not too many and certainly not too many
"hard" ones.
Definition of "hard kanji": Kanji you don't know.
--
Always be sincere, but never be serious.
Allan Watts
Sean
> Apud Chris Kern <chris...@gmail.com> (sci.lang.japan) hoc legimus:
>> ... I don't know where these mythical "all kana" texts
>> people keep talking about come from ....
>
> Bookshops. That's where I bought "こいぬのくんくん" by "ディック．
> ブルーナ". They're for pre-schoolers.
I think what Chris was referring to is the fact that such text is not
one continuous stream of kana. There is a sort of spacing between
phrases scheme that is a big help.
Sean
> Chris Kern wrote:
>
>> I don't know where these mythical "all kana" texts people keep
>> talking about come from
>
> In addition to the places other people have mentioned to date: the
> scripts of at least some earlier video games, particularly RPGs, are
> written entirely in kana. (snip)
Back in the 80s I used to play video games at a local arcade in
Ichikawa. In fact, my future wife and I did this together. There was
one game in which boxes would open with clues for your next move. The
clues appeared only for a moment, and they were written in katakana
only. Even my wife found it next to impossible to read them before they
disappeared.
Remember telegrams? They were all in katakana. I've had the opportunity
to witness NSsoJ read telegrams aloud, stumbling over the words and
sometimes frowning at the text in perplexity. But I bet the boys down
at the telegram office could read them without a blink.
Joachim Pense
Look at it from the other side:
It sometimes happens when I read stories to my children, that my mind
wanders and I start thinking of something else, without stopping
reading. And I don't remember a bit of what I just read loudly. (Of
course I don't like this happening, but anyway.) So the reading-voice
connection can even work without involvment understanding.
Joachim
Nigel Greenwood
It sounds to me as if Japanese studies in the anglophone world may
still be awaiting their John DeFrancis. For the benefit of those who,
unlike Mike Wright, haven't studied Chinese, here is an excerpt from
the Preface to DeFrancis' Advanced Chinese Reader:
"The teaching of Chinese reading has suffered from over-emphasis on
acquisition of characters at the expense of learning many compounds and
doing extensive reading. It is easy enough to look up characters in a
dictionary, but it often happens that what they actually mean in a
given text can be grasped only if the student has achieved a sort of
intuitive feeling for combinations and contexts through extensive
involvement in the total act of reading."
I expect that these comments apply in large measure to Japanese as
well.
DeF expanded on these ideas in a 1966 article entitled "Why Johnny
can't read Chinese". (Incidentally, it's a pleasure to record that
this Nestor of Chinese studies is still active & publishing papers at
the age of 95.)
Ben Finney
> Ben Finney posted:
> >Having learned the kanji, I definitely find the kanji version of a
> >text easier to read than without. The easiest explanation I can
> >come up with is the huge number of homophones in Japanese, which
> >are directly disambiguated by the kanji meanings.
>
> It's just because you're used to reading the kanji-kana mix.
When I knew only kana, it was laborious to read a passage of only
kana. This was because I had to "sound out" every word while reading
it, which was very slow. I was used to only kana, not kanji at all, so
your explanation doesn't apply.
When I began learning kanji, I stopped "sounding out" words, and was
able to recognise words by their *visual appearance*. This is very
similar to why I can read English text quickly: I recognise word
shapes, rather than sounding out letters.
There has never been a stage where I find it easy to read kana-only
texts.
> >In spoken language, of course, the meanings of homophones are
> >indirectly disambiguated by context (accents, situational,
> >non-spoken communication, etc.)
>
> I'm curious why you think that context does not exist in written
> language as well.
It certainly exists, but all the context that exists in a passage of
printed text is also present in in-person speech; and speech benefits
from much *more * contextual information.
> Are you honestly saying that if a native Japanese person reads some
> written material out loud without making any gestures, another
> native speaker will be unable to understand it because of
> homophones?
I'm not making a claim of "unable to understand"; I'm claiming "much
more tedious to read".
To be equivalent, the reading of the text would need to be mechanical
(as the transmission of syllables in printed text is mechanical), as
for example by a robotic speech synthesis unit, not read aloud by a
human.
> Can you give some examples of *actual* sentences using these
> homophones that cannot be disambiguated via context (i.e. not
> sentences purposely constructed like this).
You're right that the burden is on me for making the positive
assertion. I will try to do so when I have some more time.
--
\ "I have a large seashell collection, which I keep scattered on |
`\ the beaches all over the world. Maybe you've seen it." -- |
_o__) Steven Wright |
Ben Finney
Ben Finney
> Right, but this does not mean it's not still a sound-based process.
This appears to be an endless circuit: everything described as having
no connection with sound, you seem to respond with "sure, but that
doesn't mean it's not still based on sound". I don't see how this is
different from saying "sure, but all that doesn't mean it's not
magical" at every stage.
However, you've kindly gone into more detail on what you actually
mean:
> As I said in another post, when a linguist says that reading is a
> "sound-based process" he doesn't mean that you literally move your
> lips or hear a voice in your head, or that your reading speed is
> limited by the speed at which you can hear something. Obviously
> none of that is true. All it means is that the process of reading
> relies on the language ability that you have acquired through
> auditory means, and that as such, it's not really correct to speak
> of "bypassing sound" to somehow get directly to meaning.
Thanks, that makes it a lot clearer what you're talking about. All
you're arguing, though, is that the part of the brain that deals with
language is closely correlated with the reception of sound.
I don't see that any of this supports your initial claim in this
sub-thread: that the kanji *primarily* transmit sound, rather than
meaning.
--
\ "The Stones, I love the Stones; I can't believe they're still |
`\ doing it after all these years. I watch them whenever I can: |
_o__) Fred, Barney, ..." -- Steven Wright |
Ben Finney
Ben Bullock
"Low Zhen Lin" <lo...@hotmail.com> wrote in message
news:453ba...@news.tm.net.my...
Really? I don't think I've ever seen any writing with spaces before
particles.
I made a short page about it yesterday:
http://www.sljfaq.org/w/Wakachikaki
Here is the raw text for people who don't want to click on links:
{{nihongo|Wakachikaki|分かち書き}} is the name given to Japanese written
with spaces. {{jr|Wakachikaki}} is usually used for books for small children
who have not yet studied kanji.
Unlike the romanized Japanese used in textbooks, the Japanese system of
breaking words with spaces usually treats particles as part of the preceding
word, and also treats combinations using the [[possessive no]] as one word.
[[category:writing]]
Comments, ideas, etc. welcome.
I'll be uploading an image to illustrate when I have time.
--
sci.lang.japan FAQ: http://www.sljfaq.org/afaq/afaq.html
Ben Bullock
"Sean" <se...@fake.com> wrote in message
news:2006102212450475249-sean@fakecom...
> On 2006-10-22 02:45:32 -0700, Ben Finney
> <bignose+h...@benfinney.id.au> said:
>>
>> Yes, it would be a close call. I don't read many passages in pure
>> kana, so haven't really got any empirical (or anecdotal) evidence
>> there.
>
> I imagine there's a few us in here (Ben is one I bet) who have had a lot
> of experience reading kana-only children's books aloud to our kids when
> they were very little.
I have absolutely zero experience of it. If I'm forced to read a book in
Japanese (my children like "sora-mame-kun", if anyone knows that) I just
translate the books into English as I read. But I usually insist on reading
English books to them.
Bart Mathias
> [..]
> It sounds to me as if Japanese studies in the anglophone world may
>
> [..] (Incidentally, it's a pleasure to record that
> this Nestor of Chinese studies is still active & publishing papers at
> the age of 95.)
I hope you're right. It occurs to me that he hasn't shown up at any
Faculty Retirement Association, UH Manoa meetings in a year or so.
He was hale and hearty for this presentation, though:
Activities, 2001-2002
11/08 John DeFrancis, Emeritus Prof. of Chinese, "In the Footsteps of
Kublai Khan"
wherein he relates an adventure he undertook as a young man in the
boondocks of China and Mongolia.
Bart Mathias
> [..] I prefer kanji/kana over kana or romaji, but
> that doesn't mean my verbal skills are anything to write home about. My
> reading comprehension has always been better than my aural
> comprehension, precisely because the written version gives me better
> grammar clues than verbal Japanese.
I'm another whose reading surpasses his aural comprehension, but it's
not the grammar.
I can slow my eyes down when necessary, but not my ears. (Fortunately
back in my subtitling days, replaying was almost as good as slowing down
ears, or I would'na got rich).
Bart
Bart Mathias
> [...]
> Remember telegrams? They were all in katakana. I've had the opportunity
> to witness NSsoJ read telegrams aloud, stumbling over the words and
> sometimes frowning at the text in perplexity. But I bet the boys down at
> the telegram office could read them without a blink.
That almost reminds me of the anecdote about the telegram that read in part:
カネオクレ
and the kokoro was misreading 送れ as 遅れ or vice versa. Or maybe not.
I sure liked the Good Old Days when I could remember some things.
Bart
Kevin Wayne Williams
Rich? You're rich? And here I thought that you had Buck R. Yarrow as a
boarder because you needed the extra income.
KWW
Bart Mathias
>
> "Low Zhen Lin" <lo...@hotmail.com> wrote in message
> news:453ba...@news.tm.net.my...
>
>> Nigel Greenwood wrote:
>>
>>> On a different point, what exactly is the objection (apart from saving
>>> paper) to using gaps to separate words in kana? Granted, it's not
>>> traditional -- but then ancient Greek, & to some extent Latin, used to
>>> be written without gaps, yet modern Greeks & Italians are happy enough
>>> to use them!
>>
>>
>> Methinks the issue is not so much using spaces as it is where to put the
>> spaces. One camp will put spaces after particles, but not before, the
>> other camp will put spaces around particles.
>
>
> Really? I don't think I've ever seen any writing with spaces before
> particles.
>
> I made a short page about it yesterday:
> http://www.sljfaq.org/w/Wakachikaki
>
> Here is the raw text for people who don't want to click on links:
>
> {{nihongo|Wakachikaki|分かち書き}} is the name given to Japanese written
> with spaces. {{jr|Wakachikaki}} is usually used for books for small
> children who have not yet studied kanji.
You sure? I is allus said "wakachi*g*aki" for years and years.
Bart
Ben Bullock
"Bart Mathias" <mat...@hawaii.edu> wrote in message
news:yPidneSJa4q-kqHY...@hawaiiantel.net...
> Ben Bullock wrote:
>> {{nihongo|Wakachikaki|分かち書き}} is the name given to Japanese written
>> with spaces. {{jr|Wakachikaki}} is usually used for books for small
>> children who have not yet studied kanji.
>
> You sure? I is allus said "wakachi*g*aki" for years and years.
Oops. Thanks for the correction.
Ben Finney
> I imagine there's a few us in here (Ben is one I bet) who have had a
> lot of experience reading kana-only children's books aloud to our
> kids when they were very little.
No, I've never had such an experience.
--
\ "I was born by Caesarian section. But not so you'd notice. It's |
`\ just that when I leave a house, I go out through the window." |
_o__) -- Steven Wright |
Ben Finney
Peter T. Daniels
You sure he's only 95? I'd guess more like 99 or 100, but I haven't
talked to Jim Unger in a while (or to John DeF in much longer than
that).
j...@csse.monash.edu.au
>Nigel Greenwood wrote:
>> DeF expanded on these ideas in a 1966 article entitled "Why Johnny
>> can't read Chinese". (Incidentally, it's a pleasure to record that
>> this Nestor of Chinese studies is still active & publishing papers at
>> the age of 95.)
>You sure he's only 95? I'd guess more like 99 or 100,
The WWW CVs don't have his date-of-birth, but the University of
Pennsylvania Department of Oriental Studies published a
"Schriftfestschrift: Essays on Writing and Language in Honor of John
DeFrancis on His Eightieth Birthday" in 1991, so 95 is pretty close.
>but I haven't
>talked to Jim Unger in a while (or to John DeF in much longer than
>that).
I last spoke with Jim in 1996, but have exchanged emails more
recently. Jim, of course, is nothing like 95. He's about my age or
a bit younger.
--
Jim Breen http://www.csse.monash.edu.au/~jwb/
Clayton School of Information Technology,
Monash University, VIC 3800, Australia
ジム・ブリーン@モナシュ大学
Sean
Probably a good idea. I shamelessly used my kids as study aids.
Sean
<bignose+h...@benfinney.id.au> said:
> Sean <se...@fake.com> writes:
>
>> I imagine there's a few us in here (Ben is one I bet) who have had a
>> lot of experience reading kana-only children's books aloud to our
>> kids when they were very little.
>
> No, I've never had such an experience.
Ah yes, the other Ben.
Sean
>
> "Bart Mathias" <mat...@hawaii.edu> wrote in message
> news:yPidneSJa4q-kqHY...@hawaiiantel.net...
>> Ben Bullock wrote:
>
>>> {{nihongo|Wakachikaki|分かち書き}} is the name given to Japanese written
>>> with spaces. {{jr|Wakachikaki}} is usually used for books for small
>>> children who have not yet studied kanji.
>>
>> You sure? I is allus said "wakachi*g*aki" for years and years.
>
>
> Oops. Thanks for the correction.
Of course, there is always the slender chance that Bart was talkin
funny like a furriner for years and years, but no one had the heart or
the courage to correct him. A very slender chance.
Paul D
>>> However, it's well known ? particularly among the typesetting industry
>>> ? from analyzing people's reading behaviour that your eyes don't look
>>> at every letter when you read immersively. In fact, they don't even
>>> look at every word! Your eyes skip along, with the fovea (centre of
>>> vision) focusing on 5-7 spots per line of text, and the parafovea
>>> taking in the shapes of the surrounding words at once, which your brain
>>> interprets as words you know.
>>
>> Right, but this does not mean it's not still a sound-based process.
>> As I said in another post, when a linguist says that reading is a
>> "sound-based process" he doesn't mean that you literally move your
>> lips or hear a voice in your head, or that your reading speed is
>> limited by the speed at which you can hear something. Obviously none
>> of that is true. All it means is that the process of reading relies
>> on the language ability that you have acquired through auditory means,
>> and that as such, it's not really correct to speak of "bypassing
>> sound" to somehow get directly to meaning.
>
> I read pretty fast in English, but I still sort of hear the words in my
> head. In fact, if I don't get that feeling, I know that I'm not really
> taking in what I'm reading. (To give you an idea of how much I read,
> the summer after I graduated from high school, I averaged a bit over
> 3.5 books--mostly novels--a day for a period of about 75 days.)
Yes, almost everyone hears words in their heads as they read, but this
is because your brain already knows the pronunciation of a word, and
not because the word was written phonetically. You see the number 190
and think "one hundred and ninety", but there's nothing phonetic about
what you read. Your brain is making this connection after the fact.
Paul
Nigel Greenwood
j...@csse.monash.edu.au wrote:
> Apud Peter T. Daniels <gram...@verizon.net> (sci.lang.japan) hoc legimus:
> >Nigel Greenwood wrote:
> >> DeF expanded on these ideas in a 1966 article entitled "Why Johnny
> >> can't read Chinese". (Incidentally, it's a pleasure to record that
> >> this Nestor of Chinese studies is still active & publishing papers at
> >> the age of 95.)
>
> >You sure he's only 95? I'd guess more like 99 or 100,
>
> The WWW CVs don't have his date-of-birth, but the University of
> Pennsylvania Department of Oriental Studies published a
> "Schriftfestschrift: Essays on Writing and Language in Honor of John
> DeFrancis on His Eightieth Birthday" in 1991, so 95 is pretty close.
Maybe I'm just a better googler than you guys ... I found a CV that
gave his date of birth as 1911 -- before the First World War, for
chrissake! And I know that he recently published a very perceptive
paper on The Prospects for Chinese Writing Reform. See:
http://pinyin.info/readings/defrancis/chinese_writing_reform.html
One just hopes that, in the good old Confucian tradition, he gets more
& more respect as he gets older.
Chris Kern
<gram...@verizon.net> posted the following:
>You sure he's only 95? I'd guess more like 99 or 100, but I haven't
>talked to Jim Unger in a while (or to John DeF in much longer than
>that).
Actually I'm taking a class from Professor Unger right now (on east
asian scripts, go figure) and he said about 95 for DeFrancis.
-Chris
Ben Bullock
> On 2006-10-22 18:02:59 -0700, Ben Finney
> <bignose+h...@benfinney.id.au> said:
>
>> Sean <se...@fake.com> writes:
>>
>>> I imagine there's a few us in here (Ben is one I bet) who have had a
>>> lot of experience reading kana-only children's books aloud to our
>>> kids when they were very little.
>>
>> No, I've never had such an experience.
>
> Ah yes, the other Ben.
There's also Mr Monroe.
Tak To
> On Sat, 21 Oct 2006 17:54:56 -0400, Tak To <ta...@alum.mit.edu.->
> posted the following:
>
>>The study said only that readers' eye movement "got stuck", which I
>>take to mean they hesitated or paused at the unfamiliar combination
>>of kanjis.
>
> Yes, but they did this significantly more often with the
> meaning-switch mistakes than they did with the sound-switch mistakes.
I am not addressing the issue of frequency, I am saying that
hestitation or pause does not mean "having trouble".
>>In all likelihood they treated it
>>as a neologism, and this does not invalidate the theory that kanji
>>communicated meaning (as well as sound).
>
> If they "treated it as a neologism" and it had nothing to do with the
> sound/meaning issue, we would expect equal trouble with either type of
> mistake.
I am not sure how you arrive at this conclusion. In the homophone
case the reader either sees nothing wrong or treats it as an error(*).
It is not possible to treat it as a neologism. And treating it as
a neologism is not "having trouble" since the intended meaning is
received.
And it is quite plausible that handling a neologism could takes more
time than a simple error since it requires a deliberate decision.
(*) I can't tell from the short description of the experiment. E.g.,
I am not quite convinced that a native Japanese could fail to detect
that 日本 erroneously written as 二本. Obviously the familiarity of
the kanji phrase is a key parameter here.
> Kanji definitely "communicate meaning", the question is
> whether they somehow communicate meaning independent of sound. This
> experiment would seem to suggest that for fluent native speakers, the
> sound value of the characters is much more important to reading than
> the meanings.
I have a small problem with the wording of the conclusion, esp.
the phrase "than the (value of) meaning" since the meaning
obtained indirectly through sound is still the meaning of the
kanji. Otherwise, it would not be possible to recognized any
error in the homophonous case. It would perhaps be clearer to
contrast "visual->sound->meaning" vs "visual->meaning" than
"kanji->sound" vs "kanji->meaning".
I have a bigger problem with the logic of the conclusion.
The experiment is not designed to recognize positive cases,
let alone the break down along "visual->sound->meaning" vs
"visual-> meaning". Thus it is is not possible to say that
"v->s->m" is used more often than "v->m".
As for the negative case with homophones, it is not clear
what portion of the errors are actually detected. As pointed
out before, detecting a homophonous error requires "v->m".
As things are, I am not sure what conclusion can be drawn
from the experiment other than the fact that "v->s->m"
is used in some cases. But that is hardly news and no
one has seriously doubted it.
Tak
--
----------------------------------------------------------------+-----
Tak To ta...@alum.mit.eduxx
--------------------------------------------------------------------^^
[taode takto ~{LU5B~}] NB: trim the xx to get my real email addr
Tak To
> This is often one of the great failings of Japanese language programs
> -- "reading" is equated with learning characters, and so "mastering
> the kana" becomes simply a factor of whether you can associate the
> kana with the correct roman letters, and "learning a kanji" means
> being able to write it, know the readings, and know the "meaning" that
> the textbook has assigned to it. Whether you can actually read the
> characters in context is only of secondary importance.
I am not sure it should be considered a failing if such programs
meet the demands of the society (both the daily life kind and the
high culture kind).
Kevin Wayne Williams
> Chris Kern wrote:
>> This is often one of the great failings of Japanese language programs
>> -- "reading" is equated with learning characters, and so "mastering
>> the kana" becomes simply a factor of whether you can associate the
>> kana with the correct roman letters, and "learning a kanji" means
>> being able to write it, know the readings, and know the "meaning" that
>> the textbook has assigned to it. Whether you can actually read the
>> characters in context is only of secondary importance.
>
> I am not sure it should be considered a failing if such programs
> meet the demands of the society (both the daily life kind and the
> high culture kind).
better than I can, could tell me little stories about the entire jouyou
kanji, complete with etymological information, but couldn't sit and
actually read a magazine article. What use is that?
KWW
Tak To
> It sounds to me as if Japanese studies in the anglophone world may
> still be awaiting their John DeFrancis. For the benefit of those who,
> unlike Mike Wright, haven't studied Chinese, here is an excerpt from
> the Preface to DeFrancis' Advanced Chinese Reader:
>
> "The teaching of Chinese reading has suffered from over-emphasis on
> acquisition of characters at the expense of learning many compounds and
> doing extensive reading. It is easy enough to look up characters in a
> dictionary, but it often happens that what they actually mean in a
> given text can be grasped only if the student has achieved a sort of
> intuitive feeling for combinations and contexts through extensive
> involvement in the total act of reading."
>
> I expect that these comments apply in large measure to Japanese as
> well.
Is he talking about teaching Chinese as L1 or L2? In any case,
DeFrancis is hardly an acknowledged expert of Chinese language
pedagogy.
If he is talking about Chinese as L1, then I don't really agree
with him. It is well known that there is a minimum core of
characters that a child must acquire before he can do any real
reading and that set should be taught as early as possible.
Note that for a young Chinese child, "intuitive feeling for
combinations and contexts" has already been achieved by the
time he starts to talk.
Bart Mathias
It was temporary. In 1985 it bought me my first trip to Japan in 28 years.
(Then _The Complete Guide ..._ got me rich again in 1991, enough to
visit my daughter in Japan and go to Malaysia, Singapore, and Korea. I
even lived in Bangkok for an hour! Alas, I can no longer leave the
state no matter how rich I get.)
Bart
Chris Kern
posted the following:
>Chris Kern wrote:
>> On Sat, 21 Oct 2006 17:54:56 -0400, Tak To <ta...@alum.mit.edu.->
>> posted the following:
>>
>>>The study said only that readers' eye movement "got stuck", which I
>>>take to mean they hesitated or paused at the unfamiliar combination
>>>of kanjis.
>>
>> Yes, but they did this significantly more often with the
>> meaning-switch mistakes than they did with the sound-switch mistakes.
>
>I am not addressing the issue of frequency, I am saying that
>hestitation or pause does not mean "having trouble".
No. But the experiment measure eye movement, such as when the eyes
had to go back and read a word again.
>(*) I can't tell from the short description of the experiment. E.g.,
>I am not quite convinced that a native Japanese could fail to detect
>that 日本 erroneously written as 二本. Obviously the familiarity of
>the kanji phrase is a key parameter here.
There is no "obviously" or "I am not quite convinced" here -- it's
experimentation and data. You may not like the fact that a native
speaker would not notice that mistake, but that's what the experiment
showed.
>I have a bigger problem with the logic of the conclusion.
Without even having read the details?
>The experiment is not designed to recognize positive cases,
>let alone the break down along "visual->sound->meaning" vs
>"visual-> meaning". Thus it is is not possible to say that
>"v->s->m" is used more often than "v->m".
>
>As for the negative case with homophones, it is not clear
>what portion of the errors are actually detected. As pointed
>out before, detecting a homophonous error requires "v->m".
>
>As things are, I am not sure what conclusion can be drawn
>from the experiment other than the fact that "v->s->m"
>is used in some cases. But that is hardly news and no
>one has seriously doubted it.
Well, go ahead and write an article refuting it, then -- you should be
able to get it published if the original experiment was so flawed and
problematic that someone can challenge it without having even read the
details.
-Chris
Chris Kern
posted the following:
>Nigel Greenwood wrote:
>
>> It sounds to me as if Japanese studies in the anglophone world may
>> still be awaiting their John DeFrancis. For the benefit of those who,
>> unlike Mike Wright, haven't studied Chinese, here is an excerpt from
>> the Preface to DeFrancis' Advanced Chinese Reader:
>>
>> "The teaching of Chinese reading has suffered from over-emphasis on
>> acquisition of characters at the expense of learning many compounds and
>> doing extensive reading. It is easy enough to look up characters in a
>> dictionary, but it often happens that what they actually mean in a
>> given text can be grasped only if the student has achieved a sort of
>> intuitive feeling for combinations and contexts through extensive
>> involvement in the total act of reading."
>>
>> I expect that these comments apply in large measure to Japanese as
>> well.
>
>Is he talking about teaching Chinese as L1 or L2? In any case,
>DeFrancis is hardly an acknowledged expert of Chinese language
>pedagogy.
He's talking about L2. And the second sentence is laughable -- who
doesn't consider him an expert of Chinese language pedagogy? His
language texts were some of the most widely used texts (they still are
used in some places, and likely would be used even more if a newer
edition of the whole set came out).
If you're talking about teaching Chinese reading to Chinese natives,
then you might be right, but JDF isn't known for that sort of
pedagogy.
>If he is talking about Chinese as L1, then I don't really agree
>with him. It is well known that there is a minimum core of
>characters that a child must acquire before he can do any real
>reading and that set should be taught as early as possible.
>Note that for a young Chinese child, "intuitive feeling for
>combinations and contexts" has already been achieved by the
>time he starts to talk.
Right, that's the big difference. The reason you can't teach reading
to a foreigner the same way you can to a native speaker is that the
native speaker is already fluent in the language and just needs to
learn how the words he already knows are represented in characters.
The example sentences that are found in readers for children in both
Japan and China are ones that the children will be completely familiar
with (although some vocabulary building may be done as well).
-Chris
Ben Finney
> What textbooks did you people use? JFPB doesn't use any kanji at all
> until the end of book 2, and never gets to the point where the text
> looks at all normal.
My teachers (and, thus, I) use _Minna no Nihongo_, in the edition that
uses natural writing with furigana throughout.
The student who doesn't want to learn kanji has kana all the way
through, but the kanji are there as an extra channel of information
and they probably can't avoid absorbing some of it.
The student who does want to learn kanji won't learn it from this
course, but can make full use of whatever kanji knowledge they are
gaining through other means.
--
\ "I have an answering machine in my car. It says, 'I'm home now. |
`\ But leave a message and I'll call when I'm out.'" -- Steven |
_o__) Wright |
Ben Finney
Rob Freeman
physicist Richard Feynman. I think it was in "The Pleasure of Finding
Things Out."
As I recall he tells of an "experiment" where he found he could
accurately time a minute in his head while reading, but not while
talking. He was sounding the numbers out in his head, and if he talked
he lost the rhythm. Then he found a friend who said he could time a
minute while talking and proceeded to demonstrate. Further examination
proved the friend could not match the time accurately while reading
(which Feynman could!) It turned out the friend was timing not by
"sounding the numbers out", but by "watching" them...
It wasn't the "sounding out" or the "watching" which was important, but
it had to be one or the other :-)
-Rob
Ben Finney wrote:
> Chris Kern <chris...@gmail.com> writes:
>
> > Right, but this does not mean it's not still a sound-based process.
>
> This appears to be an endless circuit: everything described as having
> no connection with sound, you seem to respond with "sure, but that
> doesn't mean it's not still based on sound". I don't see how this is
> different from saying "sure, but all that doesn't mean it's not
> magical" at every stage.
>
> However, you've kindly gone into more detail on what you actually
> mean:
>
> > As I said in another post, when a linguist says that reading is a
> > "sound-based process" he doesn't mean that you literally move your
> > lips or hear a voice in your head, or that your reading speed is
> > limited by the speed at which you can hear something. Obviously
> > none of that is true. All it means is that the process of reading
> > relies on the language ability that you have acquired through
> > auditory means, and that as such, it's not really correct to speak
> > of "bypassing sound" to somehow get directly to meaning.
>
> Thanks, that makes it a lot clearer what you're talking about. All
> you're arguing, though, is that the part of the brain that deals with
> language is closely correlated with the reception of sound.
>
> I don't see that any of this supports your initial claim in this
> sub-thread: that the kanji *primarily* transmit sound, rather than
> meaning.
>
> --
> \ "The Stones, I love the Stones; I can't believe they're still |
> `\ doing it after all these years. I watch them whenever I can: |
> _o__) Fred, Barney, ..." -- Steven Wright |
> Ben Finney
Mike Wright
This seems right. (Except that I thought "one-ninety".)
How would a Japanese register sound just by looking at a kanji without
taking the context into account?
For that matter, I wonder what everyone will hear in their heads when
they see: "He had the lead."
Which does it go with?
"Bob was in a play last week."
"I had the bullet molds, but I had to wait for Bob."
--
Mike Wright
http://www.raccoonbend.com
Mike Wright
> On Mon, 23 Oct 2006 17:18:31 -0400, Tak To <ta...@alum.mit.edu.->
> posted the following:
>
>>Is he talking about teaching Chinese as L1 or L2? In any case,
>>DeFrancis is hardly an acknowledged expert of Chinese language
>>pedagogy.
>
> He's talking about L2. And the second sentence is laughable -- who
> doesn't consider him an expert of Chinese language pedagogy? His
> language texts were some of the most widely used texts (they still are
> used in some places, and likely would be used even more if a newer
> edition of the whole set came out).
Yep, I've used them myself, and can't imagine any better. Their only
failing is that they are no longer close enough to the modern language.
As far as methodology goes, they're still great.
Chris Kern
posted the following:
>Yes, almost everyone hears words in their heads as they read, but this
>is because your brain already knows the pronunciation of a word, and
>not because the word was written phonetically. You see the number 190
>and think "one hundred and ninety", but there's nothing phonetic about
>what you read. Your brain is making this connection after the fact.
Right, I think this is what I'm trying to say -- the written symbols
make the connection with the pronunciation you already know. In a
sense no writing system is truly phonetic, but Chinese characters have
strong phonetic connections even if they're not a syllabery or an
alphabet. For instance, I think most experienced Japanese learners or
native speakers could guess the reading of 馼 even without any
context.
-Chris
Geoff
> On 20 Oct 2006 03:55:31 -0700, "Nigel Greenwood"
> <ndsg...@yahoo.co.uk> posted the following:
>
>> Is this method of skimming widely used?
>
> Possibly, but it's not very useful or successful. It's based on the
> long-standing myth that Chinese characters represent ideas rather than
> sounds and that somehow meaning can be extracted from them even if you
> don't know the words in the language. Pure nonsense, but one that
> shows up even in textbooks authored by language teachers.
>
> -Chris
Myth? Not completely. You do not need to know anything about the
pronunciation of Chinese characters to read and understand them.
Peter T. Daniels
Nigel Greenwood wrote:
> j...@csse.monash.edu.au wrote:
> > Apud Peter T. Daniels <gram...@verizon.net> (sci.lang.japan) hoc legimus:
> > >Nigel Greenwood wrote:
> > >> DeF expanded on these ideas in a 1966 article entitled "Why Johnny
> > >> can't read Chinese". (Incidentally, it's a pleasure to record that
> > >> this Nestor of Chinese studies is still active & publishing papers at
> > >> the age of 95.)
> >
> > >You sure he's only 95? I'd guess more like 99 or 100,
> >
> > The WWW CVs don't have his date-of-birth, but the University of
> > Pennsylvania Department of Oriental Studies published a
> > "Schriftfestschrift: Essays on Writing and Language in Honor of John
> > DeFrancis on His Eightieth Birthday" in 1991, so 95 is pretty close.
>
> Maybe I'm just a better googler than you guys ... I found a CV that
> gave his date of birth as 1911 -- before the First World War, for
You must be a lot younger than I've thought; I've known many, many
people who were born in the 19th century (such as my grandmother). (And
her two older sisters.)
That was even before the Boer War! (Or the Spanish-American.)
Peter T. Daniels
Something felonious in your recent past?
Brian M. Scott
<gram...@verizon.net> wrote in
<news:1161660868.6...@m73g2000cwd.googlegroups.com>
in sci.lang,sci.lang.japan:
> Nigel Greenwood wrote:
[...]
>> Maybe I'm just a better googler than you guys ... I
>> found a CV that gave his date of birth as 1911 -- before
>> the First World War, for
> You must be a lot younger than I've thought; I've known
> many, many people who were born in the 19th century (such
> as my grandmother). (And her two older sisters.)
On the other hand, I'm a couple of years older than you, and
I've not really known anyone who was born in the 19th
century, though I've met a very few. I did meet two of my
grandparents, though only once each, in both cases before I
was eight, but those are the only ones of which I'm certain.
[...]
Brian
Mike Wright
I'm very skeptical about this statement--perhaps because I've never
learned a character without learning at least one pronunciation fot
it--and I often forget meanings before pronunciation, but never, AFAIK,
the other way around.
While it's not necessary to know any particular standard pronunciation,
if I were shown a character and told that it means "gooseberry cobbler",
I'm sure that whenever I saw it, I would read it as "gooseberry
cobbler". I'm not sure what I'd do if I were told that it "indicates
that the action represented by the preceding verb is complete".
I'm trying to imagine teaching Mandarin sentence patterns written in
characters and not providing any clue as to pronunciation. Even
translations would give the students sound-related hooks for most
characters, leaving only the grammatical particles with long-winded
explanations truly unpronounceable.
How do people who are deaf from birth manage to learn to read? What kind
of internal representation(s) do they have of the morphemes of the
written langauge? Do they identify them with equivalents in sign?
If they have neither spoken language nor sign, can they actually learn
to read?
Joachim Pense
>
> Myth? Not completely. You do not need to know anything about the
> pronunciation of Chinese characters to read and understand them.
Isn't that true for every writing system?
Joachim
Sean
<kww.n...@verizon.nut> said:
> They don't. I've met advanced Japanese students who could speak much
> better than I can, could tell me little stories about the entire jouyou
> kanji, complete with etymological information, but couldn't sit and
> actually read a magazine article. What use is that?
> KWW
The speaking could come in handy. I don't think many people have
chatted up a girl by silently reading a magazine at them.
Sean
> It was temporary. In 1985 it bought me my first trip to Japan in 28 years.
>
> (Then _The Complete Guide ..._ got me rich again in 1991, enough to
> visit my daughter in Japan and go to Malaysia, Singapore, and Korea. I
> even lived in Bangkok for an hour! Alas, I can no longer leave the
> state no matter how rich I get.)
>
> Bart
Trapped in Hawaii. That's a tough one, Doc.