|I had thought that The Genesis Site had shut down, but it now
seems that it has just moved, so I've updated all links.
It also seems that it is still based on The Discovery of Genesis, in spite of the fact that Dr. Ethel Nelson, the chief author of that work, has stated that the book was "not accurate" and that she "dissuade(s) readers from referring to it".
I've decided, however, to leave these pages up for a while, since not everyone has gotten the word, and people continue to refer to claims that appeared on that site.
Over the past several years, Dr. Nelson, along with various collaborators, has continued to take a basically flawed approach to the study of the origin of Chinese characters, treating most as relatively complex semantic composites. For those who have read any of those later works and been tempted to believe the stories presented in them, I suggest two scholarly books, written by professionals in the fields of Chinese language and linguistics. If reality is more interesting to you than fantasy, don't take my word for it--read them for yourself.
1. The Origin and Early Development of the Chinese Writing System (American Oriental Series, Volume 78), by William G. Boltz, American Oriental Society, New Haven, CT, 2003. (First published in 1994.)
This book begins with basics, like the shift from proto-writing to true writing, including developmental features and stages shared by the Chinese, Egyptian, Mayan, and Mesopotamian (Sumerian/Akkadian) scripts. A major theme, and one that is most damaging to the Nelson approach, is the highly phonetic nature of even the earliest stages of the Chinese writing system.
Here's a description from a footnote in the following book: "... Boltz's work is fairly theoretical and calls for the reader to consider the relationship between sound and meaning via graphic forms from the vantage point of the development of writing in general. About 400 Chinese characters are discussed either as words or as graphs to illustrate important hypotheses concerning the theme, as given in the title of his book."
2. Chinese Writing, by Qiu Xigui (Eary China Special Monograph Series No. 4, translated into English by Gilbert L. Mattos and Jerry Norman), The Society for the Study of Early China and The Institute of East Asian Studies, University of California, Berkely, 2000.
This one has a somewhat different focus. Here's a continuation of the footnote quoted above: "By contrast [to Boltz's work], the present work is mostly descriptive, covering the origin, structure, and use of individual characters from the Shang period down to the present day. It discusses 2,370 Chinese characters and, therefore, can be used even as an extended commentary to dictionary entries, particulary in grammatological aspects."
For now, I will respond specifically to those claims that have been
at The Genesis Site. (My purpose is not to write a
on the origins of Chinese characters, but simply to show where the
of The Genesis Site have gone wrong.)
I want to say at the beginning that I welcome any substantiated corrections--citing sources where appropriate. Send emails to mike at this domain (raccoonbend, with the usual dot suffix).
Mandarin pronunciations will be given in Pinyin romanization, with tone numbers, rather than tone marks. Reconstructed Old or Middle Chinese pronunciations, if any come up, will use ASCII IPA symbols.
The format of the following pages will be a link to an original point of discussion on The Genesis Site, followed first by an illustration and then by an explanation from Chinese Characters: Their origin, etymology, history, classification and signification, by Dr. L. Wieger, S.J. This book (hereafter referred to as "Chinese Characters") was originally published in French in 1915. The English version was translated by L. Davrout, S.J., and was published in 1927 by the Catholic Mission Press. Chinese Characters is based on a combination of Chinese analysis from the Han Dynasty dictionary, Shuo1wen2 Jie3zi4, and Wieger's own scholarship. The core of Chinese Characters is a set of 177 lessons, showing various "primitives" (characters that stand on their own), phonetic elements, and compound characters. Characters are shown in modern brush-written style and in the older seal style. (The Dover paperback reprint of Chinese Characters, ISBN 0-486-21321-8, is inexpensive and very well-bound.) Due to the extensive mixing of English and Chinese in the explanatory text, I have made some slight modifications to the text, making it all English with some romanized Chinese, but I don't think that I have done anything that will change the meanings of the text.
It is important to note that modern brush and printed forms of characters are often quite different from the original forms. The pointed hair brush was a very convenient invention, but it is not suitable for rendering the complex seal characters with their wavy lines and curves of uniform thickness. As characters were adapted to the brush, the shapes of various elements were often changed in very extreme ways. Furthermore, elements that were different in the originals often came to be written the same with the brush, while elements that were the same in the originals often came to be written differently with the brush. As a result, it is pretty much impossible to analyze a character on the basis of its modern form.
Another important point is that the vast majority of Chinese
characters are compounds that include a phonetic element, along with a
semantic element (also referred to as a "signific" or "radical"). On
page 84 of The Chinese Language: Fact and Fantasy (University
of Hawaii Press, 1984, ISBN 0-8248-1068-6),
John DeFrancis provides a table (Table 3 Structural Classification of
with the following information on the proportion of characters that
a phonetic element:
|Pictographic||227 (23%)||364 (4%)||608 (3%)|
|20 (2%)||125 (1%)||107 (1%)|
|396 (41%)||1,167 (13%)||740 (3%)|
|334 (34%)||7,697 (82%)||21,810 (93%)||47,141 (97%)|
The total for the pictographic, simple indicative, and compound indicative compounds for Kang Xi is a mere 3% (approximately 1,500). Also, note that the DeFrancis book was published in 1984. Since that time the phonetic nature of even larger percentage of the Oracle Bone characters has been demonstrated.
A typical error made on The Genesis Site is to analyze a semantic-phonetic compound as though it were compound-indicative (which they refer to as "ideographic"). This will be pointed out where appropriate. This is an easy mistake for beginners to make, as a semantic-phonetic compound may not be a precise homonym of the phonetic element of that compound. The spoken languages (and Mandarin in particular) have changed a great deal over the centuries, so that, in many cases, morphemes that were once near homophones have diverged considerably. We can get hints of the original pronunciations from some of the more conservative Chinese languages, like Hokkien, Hakka, and Cantonese. Even Sinoxenic loans in Japanese, Korean, and Vietnamese, though relatively modern compared with the origins of most characters, can show us something about the earlier pronunciations. There is an entire field of phonological reconstruction of Middle and Old Chinese, pioneered by Bernhard Karlgren and carried on by scholars like Edwin G. Pulleyblank, that uses information from ancient rhyme tables, poetry, and dictionaries. Without this kind of background knowledge, how can amateurs hope to recognize phonetic elements when modern Mandarin has changed so much?
One often sees comments that Chinese Characters is outdated and inaccurate. This is because even Wieger (and his Chinese sources) often analyzed semantic-phonetic characters as though they were compound indicative. Advances in phonological reconstruction techniques, and the constant increase in the knowledge base, have gradually revealed that many more characters are of the semantic-phonetic type than was previously realized. I have begun, for the most part, with references straight from Wieger, but I intend to gradually upgrade the information to correct any such errors. I am not worried about this particular shortcoming in the context of this Web site, because any corrections will mean that the explanations given on The Genesis Site are even farther off than they appear to be on the basis of Wieger's analyses.
Because the authors of The Genesis Site rely exclusively on the modern character forms, they sometimes make assumptions about the elements of a character that cannot be justified by its original form. This often leads to analyzing what was originally a single element as a compound of two or more different elements. This is why I have included the older forms, along with Wieger's analysis.
Wieger uses the "small seal" characters, but the oldest are the "oracle bone" characters. To see some of these, click here. Click the "Next" button to see more. To see a large bone with translation, click here. For a brief overview of the stages of Chinese writing, click here.
In fact, The Genesis Site is based on The Discovery of Genesis, by C.H. Kang and Ethel R. Nelson, published in 1979. After The Discovery of Genesis was criticized for this flaw, Nelson wrote a new book, Genesis and the Mystery Confucius Couldn't Solve, with Richard E. Broadberry, which does use older character forms (though this later book has its own flaws). I will note those instances where Nelson and Broadberry no longer support the original analysis of a character. Out of 43 major points made on The Genesis Site, fully 31 of those are no longer supported in the later book.
In some cases, Nelson and Broadberry make similar claims, but revise
details to fit the older character forms. This shows the "Just So
Stories" character of the entire enterprise. Nelson and her
collaborators began by
deciding that Chinese characters must somehow reveal a knowledge of
on the part of the ancient Chinese. They then try to find pieces that
be made to fit their preconceptions. There is no way that such an
can fail. It's like making interpretations of the writings of
I'm sure that one can just as well prove that the ancient Chinese knew
advance of the teachings of Islam and Buddhism, as well as the outcome
the American Civil War.
A typical example is the character for lai2, meaning "to come". The modern form of the character looks like the character for tree (mu4) with two smaller characters for "person" (ren2) under it. Even if it really were composed of those elements (which it is not), it's still quite a stretch to say, "Oh, a tree. It must be the Tree of the Knowledge of Good and Evil. Oh, two people. They must be Adam and Eve. Well, then, if this character represents Adam and Eve coming out from behind the Tree of the Knowledge of Good and Evil, then the ancient Chinese obviously knew about the story of Genesis." This way of basing a conclusion on a series of unfounded assumptions is the key methodolgy of both books--in fact, the only methodology. While Genesis and the Mystery Confucius Couldn't Solve may use earlier character forms than does The Discovery of Genesis, there has been no improvement in methodology. (As for lai2, it was originally a pictograph of a stalk of wheat, as can be seen both from its earliest graphical forms and from its use in actual ancient texts. It's use to mean "to come" is an excellent example of a phonetic loan. How can someone who doesn't know about the early forms and the earliest use of a character hope to analyze its origin?)
An early error on The Genesis Site is the listing of the characters for the numbers 1 through 9 as pictographic, whereas they are actually simple indicative (though, I suppose. the character for "six" might count as compound indicative).
Here are Wieger's explanations of the characters for 4, 6, and 8, which will be relevant to later discussions:
Si4. Four. Numerical sign. Even number, which is easily divided into two halves. The old form graphically represents the division of si4 into two halves.
Liu4. Six. The even number, also easily divisible, that comes after four. Si4 marked with a dot. Note that in the other simple even numbers, the divisibility is also indicated; er2 two; ba1 eight.
Ba1. Etymological sense, to divide, to partake. It is a primitive representing the division in two parts, the separation; Bie2 ye3. Xiang1. Fen4 bie2 zhi1 xing4. This character now means "eight", this number being easily divided into two equal parts (note that four, a square, is a kind of unity in the Chinese reckoning).--It is the 12th radical, phonetic series 8.--In the compounds, ba1 placed on the top of the character, is sometimes reduced to two points in the modern writing.... Most of the characters having ba1 at the bottom in the Kang Xi dictionary...have really nothing in common with this primitive.
I have my own hypothesis that ba1 in the meaning of "eight" is actually a phonetic loan, and that the original morpheme came to be written with the character that was being used for bie2 (to divide, to separate), which then required that a new character be created for bie2. I'll try to add some evidence for this when I get a chance. For now, note that Pulleyblank's Lexicon of Reconstructed Pronunciation in Early Middle Chinese, Late Middle Chinese, and Early Mandarin shows the Early Middle Chinese pronunciation of [p@it]/[pE:t] for ba1 and [piat] for bie2. These are still close enough for a phonetic loan, although I haven't yet found any reconstructions of the Old Chinese pronunciations of these morphemes. If I'm correct, the fact that eight is an even number is probably irrelevant.
In order to avoid placing too many graphics on a single page, I have set up one page corresponding to each of the following pages from The Genesis Site:
The Creation The Garden of Eden The Fall The Flood The Tower of Babel
Back to RaccoonBend.com Home Page