A Theory of the Origin of Chinese Writing

Author(s): Berthold Laufer

Source: American Anthropologist, New Series, Vol. 9, No. 3 (Jul. - Sep., 1907), pp. 487-492
Published by: Wiley on behalf of the American Anthropological Association
Stable URL: https://www.jstor.org/stable/659533
Accessed: 27-02-2019 09:10 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

American Anthropological Association, Wiley are collaborating with JSTOR to digitize,

preserve and extend access to American Anthropologist

This content downloaded from 128.199.167.194 on Wed, 27 Feb 2019 09:10:01 UTC
All use subject to https://about.jstor.org/terms
A THEORY OF THE ORIGIN OF CHINESE WRITING'

By BERTHOLD LAUFER

It is not my purpose in this paper to initiate the

mysteries of Chinese writing, nor to present a feat
erudition. I merely wish to illustrate the applicatio
derived from the investigation of primitive ornam
question of the origin of ancient Chinese writing.
Every casual observer will be impressed by the d
mental and picturesque feature of Chinese chara
observation coincides perfectly with the view held
themselves, that writing is an art - a decorative art
eagerly aspired to, and occupies the same high ra
The art of painting itself received a strong impetu
penmanship, and is still markedly graphic in ch
famous painters have at the same time been note
and their autographs, one or two words dashed off w
of the brush, excite as much admiration and are as
as their sketches or water-colors. Writing, consequ
the first field for the practice of art : it was the beginn
and painting; hence in view of this fact we are justif
ing its claims, from the anthropological viewpoint,
ment of decorative art.

For such a study we must entirely eliminate the modern forms

of characters, which have been in use for two thousand years, and
turn to the oldest existing specimens of writing, which are handed
down on the bronzes of the Shang dynasty, dating from the third
millennium before Christ. At that early age the formation of
writing was completed; all further stages in its development are
either new combinations or simplifications and changes of form con-
ditional upon the changes in writing implements. The invention of
the writing-brush, of ink, and of rag-paper, necessarily produced a
1 Read at the meeting of the New York Academy of Sciences, March 25, 1907.
487

This content downloaded from 128.199.167.194 on Wed, 27 Feb 2019 09:10:01 UTC
All use subject to https://about.jstor.org/terms
488 AMERICAN ANTHROPOLOGIST [N. s., 9, 1907
tremendous effect on the shaping of characters, with a tendency
toward more rounded, graceful, and pleasing forms; while the
oldest writing materials--like bamboo, wood, stone, and bronze,
later on also silk--inscribed with a clumsy stylus and varnish,
certainly allowed of only rudely executed characters. From this
field an abundant supply of examples could be furnished on the
question as to how ornaments change under the influence of new
technique and material.
Another point that must appeal to the anthropologist is the fact
that the Chinese have anticipated us, dissected, analyzed, and inter-
preted all their characters in numerous philological works com-
manding high respect. From the results of their painstaking re-
search, foreign scholars have elaborated their system of writing, and
usually have adhered to the native interpretations with implicit
faith. But these interpretations, however ingenious and convincing
they may at first seem, have only a relative value as personal im-
pressions or popular traditions. Chinese scholars began with delib-
eration to reflect upon the composition and meaning of their char-
acters, and to arrange them in analytical dictionaries, as late as
post-Christian times, after writing itself had been in constant use
for at least three thousand years; so that practically they could
have known nothing about its original growth. What they have
to say concerning this point is equivalent to the oral interpretations
that we now receive from primitive tribes regarding the significa-
tion of their ornamental patterns, and must be regarded in the
same critical light. The agreement between the two phenomena is
so close that, just as different members of a tribe or of different
tribes of the same stock may ascribe to the same ornament a dif-
ferent meaning, various Chinese authors give widely varying and
sometimes contradictory explanations of the symbolism underlying
their characters; and the traditions crystallizing around them have
oscillated and also changed at times.
Chinese writing is not the result ot one and the same principle,
nor the product of one homogeneous mold; several factors have
combined toward its production, and during a period covering many
centuries. The most efficient method of construction was by means
of a large number of phonetic elements combined with ideographic

This content downloaded from 128.199.167.194 on Wed, 27 Feb 2019 09:10:01 UTC
All use subject to https://about.jstor.org/terms
LAUFER] THEORY OF THE ORIGIN OF CHINESE WRITING 489

signs. Nearly nine-tenths of all the characters now existing are

formed on the basis of this principle. If we eliminate this and
other comparatively recent developments, we come upon a group
of about six hundred simpler signs, called by the Chinese "pictures
of objects," which admit of no reduction into single components.
It is on this limited class of characters that European sinologues
have founded the theory of a pictographic origin of Chinese writing,
which, for the rest, is merely the reiteration of what the Chinese
themselves think on the subject. It is asserted that these charac-
ters, now conventionalized in drawing, abbreviated, and disfigured,
were developed from an original realistic picture portraying the
object which the character is intended to represent. It will be
readily seen that here we have the same condition of things, and
the same theory, as formerly advanced regarding the origin of prim-
itive ornament, when many conventional patterns, through the proc-
ess of evolution, were traced back to the realistic prototype from
which the pattern was named; and I am inclined toward the con-
viction that, just as we were obliged to dispel that belief, we shall
be compelled to abandon the long-cherished theory of the picto-
graphic origin of Chinese writing. Not that I would transfer
merely through analogy the results of research in primitive art to
the problem under consideration, but I wish to substantiate my belief
with the evidence accruing from this particular field, and thus cor-
roborate what has been ascertained from a study of the ornamen-
tation of modern times.

The proposition that the six hundred primitive symbols were

evolved from real pictures is not borne out by the facts, as they are
clearly laid down in the ancient bronze inscriptions of the Shang
period. Among the characters there preserved we meet with no
expression of realism, with no adequate likeness or full figure, but
only with symbols consisting of brief, sketchy, and shadowy out-
lines - conventional designs in which no sort of development from
a natural picture to a state of gradual conventionalization can be
traced. In most cases such a development would be materially
impossible and illusory at the outset. What could it signify in gen-
eral, and to primitive man in particular, to speak of reproducing a
representation true to nature - of water, river, cloud, wind, earth,

This content downloaded from 128.199.167.194 on Wed, 27 Feb 2019 09:10:01 UTC
All use subject to https://about.jstor.org/terms
490 AMERICAN ANTHROPOLOGIST [N. s., 9, 1907

metal, fire, and many others that we find among the earliest attem
of Chinese drawing? He must needs turn, with no other alter
tive, to conventional symbols to express the ideas of such obje
In fact, any realistic representations that could be construed as
ing preceded writing, and finally resulted in it, do not exist,
have never existed. They do not even exist as survivals in art,
if they ever did we should justly expect there to discover t
Ancient art, however, is in perfect harmony with ancient writ
As all primeval characters represent conventional designs, so i
early Chinese art as decidedly conventional and traditional as
art can be; and I may go a step farther by making bold to
that in the art prior to our era, illustrative of a development ex
ing over three thousand years, there is not a trace of realism
naturalism apparent in any artistic production. All pattern
either strictly geometrical or consist of animals and monsters
ventionalized to extremes, while the human figure plays hardly
conspicuous r6ole. Realism appeared in Chinese art only a
centuries after the beginning of the Christian era, in the wor
prominent individual artists, as though it were the result of a react
directed against the monotonous traditionalism of the older na
art. Not one natural bird, not a single natural tree or flower,
we discover in the archaic period, the Han dynasty included, u
in the seventh century, the great painters of lifelike bird
flowers arise in the time of the T'ang.
The opinion that conventional forms are evolved from real
representations is without substantial foundation, and is refute
far as China is concerned, by historical evidences such as thes
realism in art proves to be the product of such recent times,
difficult to imagine how it could have existed during the epoc
the embryonic formation of writing, whose beginnings must b
jectured to have been at least in the fourth millenium B.
that there is nothing left for us but to conclude that the oldest for
extant are also identical with the earliest primeval forms, whic
course had no predecessors. These forms, if we analyze them fa
are composed of a certain number of lines, strokes, dots, com
nations of these, and simple ornamental figures which are vari
interpreted as certain objects or are named after them. Ro

This content downloaded from 128.199.167.194 on Wed, 27 Feb 2019 09:10:01 UTC
All use subject to https://about.jstor.org/terms
LAUFER] THE OR Y OF THE ORIGIN OF CHINESE WRITING 49I

dots, for example, according to the different ways in which they

surrounded by lines, are identified with raindrops in the one c
with grain and rice in two other cases, and, in still other comb
nations, with sparks of fire or nuggets of metal. It will be reco
nized that it was not the picture of an object, or any attemp
draw a life-like design, that was the primary agency in the for
tion of writing, but a group of conventional ornamental for
These received individual names by which to distinguish them o
from another, the name being suggested by a process of associat
in the primitive mind, of the design with the object to which
name referred. Thus, naturally, a vertical stroke would suggest
stem of a tree or a piece of wood; a curved line, a snake or a riv
a zigzag line, the top of a mountain. This designation adhere
the ornament traditionally, and name and design finally becam
thoroughly yoked together that the symbol called to mind the na
and the name the symbol, until they became inseparably united
will not dwell at length on the final process that led to the con
tion of ornaments as true writing, in which the design was fixed
last as a character, and its name was substituted by the word co
veying the idea of the object that this name implied. This was b
no means an abstract processoof intentional rationalism, but a dev
ment as purely emotional as the original creation of ornaments
was doubtless prompted by the early existence of an elaborate sys
of ritual symbolism and by the facts that ornamental combinat
and compositions are treated as legible rebuses which have do
nated the art and religious customs of China from the days
antiquity until the present time. Whatever the psychical bas
this concluding step may have been, I think we may say now th
the beginnings of Chinese writing are not pictographic, but orn
mental and symbolic.
This theory receives strong corroboration from two other ide
graphic systems of writing occurring in eastern Asia - that of
Lolo and that of the Miaotse. Of the latter, we have a sing
specimen preserved in a Chinese book of the year 1683, giving
short songs in the original script, with an interlinear versio
Chinese. The Lolo writing, consisting of about three thous
characters, has become better known through the investigation

This content downloaded from 128.199.167.194 on Wed, 27 Feb 2019 09:10:01 UTC
All use subject to https://about.jstor.org/terms
492 AMERICAN ANTHROPOLOGIST [N. S., 9, 1907

Father Vi'al, who sees in it one of the oldest forms of Chinese

writing; while other scholars consider it as adaptations to and re-
constructions of ancient Chinese characters. Although tradition-
ally its invention is attributed to a Chinese who lived about the year
5 50, there is no resemblance whatever between Lolo and Chinese
or between Lolo and Miaotse characters. The Lolo and Miaotse

symbols are quite independent and original in their outward

ture, and no doubt originally represented indigenous ornam
those particular tribes. The stimulus of adapting these orna
designs to the purposes of writing was unquestionably re
from the Chinese, while the forms themselves were autochth
This supposition accounts as well for the above tradition as
facts as we find them at present, and in my opinion ther
other possible way of explaining them.
COLUMBIA UNIVERSITY,
NEW YORK CITY.

This content downloaded from 128.199.167.194 on Wed, 27 Feb 2019 09:10:01 UTC
All use subject to https://about.jstor.org/terms