Just for fun, I'm browsing through Basil Hall Chamberlain's "A Practical Introduction to the Study of Japanese Writing" (published in 1899). In the introduction he discusses how he came up with the 2,488 kanji chosen for the book:

"Enquiry at Tokyo printing offices then showed that the maximum number of characters employed in this country to be 9,500; but of these, over 3,000 are extremely rare, serving the needs of such writers only as affect archaic and poetical diction. The number kept on hand in all the usual varieties of size and "face" is 6,100; but this again must be regarded as a maximum, an abundantly liberal limit stretched so far only by precaution, in order to meet the multifarious requirements of commercial, legal, medical, administrative, and other technicalities, but never attained to in the practice of any one writer or even in the knowledge of the general public.

"Scholars carry over 4,000 characters in their heads, the general public about 3,000. One thousand characters, which the experience of forty years has proved to recur with special frequency, are kept by the type-foundries in larger quantities than the rest; but a few additional hundreds on the boundary line run them hard in the race, and about 1,000 more form a needful acquisition. This gives a reduced total of about 2,500 common characters with which students must familiarize themselves, whether their ultimate object in learning Japanese be mission work, diplomacy, commerce, or learned research.

"Just these indispensable characters are here brought before their notice, with explanations thrown in occasionally to ease the drudgery of memorising. A few--a very few--characters of a lower degree of usefulness may be distinguished by a keen eye among the number. But there is method even in this madness. Such characters are brought in because they help to explain others of greater importance, the total result of their introduction being to lighten the learner's task."

Chamberlain encourages the reader that it's possible to learn the required kanj: "after all, very dull Japanese boys succeed in learning the characters perfectly. Then why should we not do so?"


What I find interesting about the above is where he says that the general public knows about 3,000 kanji, because I understand that the figure today is similar (around 3,000).

"after all, very dull Japanese boys succeed in learning the characters perfectly. Then why should we not do so?"

Tell it like it is Chamberlain!

That's pretty interesting that the number is put at the same level back in the Meiji. It makes me wonder though, is that because it's a neat little number that's big, but also not too big? Is it really just a lie that's been maintained? Or does it actually reflect what you need. I wonder...

Interesting post either way, cheers.

john555 Wrote:What I find interesting about the above is where he says that the general public knows about 3,000 kanji, because I understand that the figure today is similar (around 3,000).

People come up with the same figure when they're just pulling numbers out of the air with no research behind it. The idea that even the "dullest" boy in the Meiji period knew 3000 kanji is absurd. Chamberlain probably mostly associated with the upper class in the educational system and the government, and their children all probably knew that many kanji.

The use of furigana was a lot more widespread before WW2 anyway; much popular literature used furigana on almost every kanji.

Quote:The use of furigana was a lot more widespread before WW2 anyway; much popular literature used furigana on almost every kanji.

Do you know why they stopped in most cases?

brambles Wrote:

Quote:The use of furigana was a lot more widespread before WW2 anyway; much popular literature used furigana on almost every kanji.

Do you know why they stopped in most cases?

I've heard that it's because of the Toyo Kanji list -- the original intent of the list was that publishers would confine themselves only to those kanji, and that it would become a list of all the kanji you were expected to know. Before that there was no clear indication of what you could expect a reader to know. It may have also just been a carryover from the Edo period, where popular literature was often written almost entirely in kana, with the few kanji having furigana -- this, of course, because of the relatively low literacy level even among educated readers. (In other words, it's possible that even in pre-war Showa period readers didn't need furigana as much as they used to, but it was a carryover from earlier practice.)

EDIT: I should amend that; I was thinking of earlier Edo stuff. I just looked at a woodblock of Bakin's Satomi Hakkenden and it uses a lot of kanji, but with furigana over virtually every one.

In what I read about how many characters you need(ed) to know, there's wide discrepancies.

For example, the front flap of the book A Guide To Reading Japanese which presents the 1,850 tooyoo kanji states that "Formerly the ability to read Japanese newspapers and magazines required the knowledge of no fewer than 4,000 characters. It is now possible for the student, through mastery of the characters presented in this book, to attain this ability with far less expenditure of time and effort than was previously necessary."

Yet Chamberlain (writing in 1899) implies in the quote in my original post above that learning the 2,500 kanji he presents in his book is enough. "This gives a reduced total of about 2,500 common characters with which students must familiarize themselves, whether their ultimate object in learning Japanese be mission work, diplomacy, commerce, or learned research."

That's a big difference, 4,000 vs. 2,500.

Maybe the difference is due to different ideas about what constitutes an acceptable amount of dictionary flipping. Perhaps 2,500 was enough if you didn't mind looking up lots of kanji in the dictionary whereas 4,000 was enough if you wanted to only rarely have to look kanji up.

Of course, the only way to know for sure is to check yourself (i.e., compare your ability to read newspapers now with your ability to read pre-jooyoo kanji newspapers (from microfilm perhaps)).