Chinese Character Writing

Origins and evolution of Chinese writing systems. by Lu, Wei, Aiken, Max

The history of Chinese writing extends back more than 6000 years and the Chinese writing system remains unique among all writing systems. In this paper, the origin and evolution of Chinese writing systems will be discussed. It will be shown that in the Shang dynasty (about 1200 BC) the principles of Chinese writing had been formulated and that over the following 3000 years the structure and key elements of the system have remained, although the style of writing has changed (Li, 1969; Keightly, 1989). Explanations of the possible motivation behind the invention of Chinese writing will also be discussed. It can be shown that numerals had the highest frequency of occurrence in the earliest writing system - pottery inscriptions - and this finding indicates that one of the purposes of innovation of a writing system was for counting. This paper concludes that evidence of early Chinese writing confirms there is a significant relationship between the invention of a writing system and abstract counting and accounting requirements.

Introduction: literature review on counting and Chinese writing

The rhetoric of accounting history across time and space can take heed of primitive elements of pre-history. Artefacts may send messages from the past. These messages may be revealed through an interpretive focus of relevance to the present and to interim events. An interim event of importance in Chinese accounting is the invention of writing.1 This in itself may have strong associations with counting and the need to express financial rights and obligations among citizens through accounting measurement and communication practices.

Explanation of the origin of Chinese writing had been dominated by myths and legends. Questions are often asked including: "what is the origin of writing and abstract counting?". In order to provide answers scientists and researchers may need to find archaeological evidence. Many alternative theories have been developed in order to explain the invention of writing. In the eighteenth century William Warburton (1738), Bishop of Glouscester, introduced an evolutionary theory of writing. In his book based on his observation of Egyptian, Chinese, and Aztec manuscripts, he introduced the first evolutionary theory of writing. He argued that all scripts originally developed from narrative drawings. Although some have argued his theory does not offer a perfect explanation of the origin of particular writing systems, it has remained dominant for more than two hundred years (Schmandt-Besserat, 1996).

Archaeological findings at Uruk since the 1930s have instigated criticism of the universal pictographic theory. Archaic tables excavated have contradicted Warburton's theory. This has been because there were rarely any true pictorial signs in these tables. Even with few being truly pictorial, they were uncommon. This raised a series of uncertainties about all scripts having originated from narrative drawings (Falkenstein, 1964; Schmandt-Besserat, 1996).

In the search for the origin and invention of writing several scholars have attempted to link writing, abstract counting and accounting. These include Oppenheim (1959), Falkenstein (1964) and Amiet (1966). Falkenstein argued that cuneiform writing was originally created for the exclusive purpose of recording economic transactions. Oppenheim (1959) revealed the cuneiform envelope containing tokens used for accounting purposes. The numbers of tokens inside the envelope reflect the economic transactions incised on the surface of a cuneiform envelope. Amiet (1966) interpreted clay counters in envelopes of the prehistoric period as representing commodities.

Professor Denise Schmandt-Besserat of the University of Texas, an archaeologist specialising in pre-historic clay objects, claimed that small tokens found in the Near East are among the antecedents of writing. Schmandt-Besserat (1992a; 1992b) describes clay objects as "tokens". These different shapes of objects - cones, spheres, disks, cylinders, and so on - served as "counters" in the prehistoric Near East. They were present throughout the Near East including in Israel, Syria, Iraq, Turkey and Iran. The layers range from 8000 to 3000 BC and even later. Apart from individual clay tokens which often were loosely distributed in pre-historic sites, archaeologists discovered hollow clay balls containing such tokens. The oldest of these receptacles (called envelopes by Schmandt-Besserat) dated back to about 3250 BC. They bear seals impressed on the surface.

From about 3200 BC onward the container surface not only bears a seal but also is imprinted with every token contained in the envelope. Obviously, the need to identify the contents from outside (that is, without breaking the seal and envelope) was soon realized. This imprinting of the tokens on the surface of the envelope is considered to be a device moving towards the invention of writing.

In visiting many museums as well as archaeological sites, Schmandt-Besserat puzzled over these tokens and containers. She provides the following hypothesis: that clay tokens represented various commodities. Before 3250 BC the tokens were presumably kept in perishable containers; but after this date they were preserved in clay envelopes, each representing a commodity aggregate. This was owed by one person to another or, more often, owed to a temple precinct. At the same time there existed an alternative system using the same tokens. However, these were perforated, stringed, and held together by a sealed button of clay. The debtor was identified with the seal (wrapped around the envelope or impressed on the clay connecting the ends of the strings). This system served as a counting device and an accounting method for control of goods in the pre-historic cultures of the Near East.

The envelope was further improved from 3200 BC. By this time, each token was impressed on the outside of the envelope before it was placed into the receptacle. This enabled identification of the debtor and also a quick identification of the contents without opening the envelope. Therefore, the sum total of the various tokens in a particular envelope or on a string indicated how much was lent to a debtor by a creditor.

Realizing that it was unnecessary to make markings on the outer surface of an envelope, solid tables and solid clay balls bearing markings were then used to replace the hollow envelopes filled with tokens. The markings gradually became a system of their own, a system of writing. Based on the above, Schmandt-Besserat was able to conclude that the immediate precursor of cuneiform writing was a system of tokens.

She then summarized three evolutionary phases of counting: (1) one-to-one correspondence (mainly through tallies, pebbles, and the like); (2) concrete counting (mainly with tokens); and (3) abstract counting (with numerals). The token system reflected an archaic mode of "concrete" counting. She concluded that writing is the outcome of abstract counting which is not, as previously assumed, subservient to writing.

Schmandt-Besserat was originally not searching for, but rather stumbled upon, the origin of accounting. The most revolutionary of her discoveries are: (1) accounting existed thousands of years before writing and abstract counting; and (2) accounting became the impetus through which writing (as well as counting in the abstract sense) was created. Mattessich (1987; 1991; 1994) interpreted Schmandt-Besserat's research from the accounting point of view. That is, token shapes fulfilled the function of commodity accounts (in the generic sense) and envelopes as receivables/payables accounts contained not only the details of loans but also separately (as imprints) the total equity. This equity comprises commodities loaned, or stored in specific places and allocated to specific flocks. Therefore Mattessich deduced that the ancient Sumerians practiced a kind of double-entry record keeping some 5,000 years ago.

Robinson (1995) calls "the Ice Age signs and other forms of partial writing 'proto-writing' ... . Proto-writing long preceded the emergence of full writing ..." (p.53). The two forms of proto-writing which are particularly worth mentioning and which are relevant here are counting and bookkeeping. There were clay tablets (Figure 1) which are signs. They consist of numerals and symbols that are pictographic or quasi-pictographic. On the left side of Figure 1 is the early clay tablet from Uruk. The top is the front of the clay tablet and the bottom is the back of the clay tablet. The commodities are unknown, but the calculation on both sides can be summarized as follows.

Up to now there is little evidence from other ancient civilizations such as the Egyptian and Chinese cultures. The primary purpose of this paper is to provide some elementary archaeological evidence from China to support Schmandt-Besserat's theory. The remainder of the paper is organized as follows: (1) Development of Chinese writing systems; (2) The theory of origin and evolution of Chinese writing systems; (3) Why choose inscriptions on pottery to study the relationship between counting and writing in China?; (4) Numerals among the inscriptions on tortoise shells; and (5) Conclusion.
Development of Chinese writing systems

People in different civilizations have invented different writing systems, for example, scripts in Egypt, the cuneiform images in Mesopotamia, and Chinese characters in China (Gaur, 1984). Without such inventions, it is difficult to imagine how human beings could have achieved today's prosperity. Robinson (1995, p.7) notes:

Writing is among the greatest inventions in human history, perhaps the greatest invention, since it made history possible. Yet it is a skill most writers take for granted. We learn it at school, building on the alphabet, or (if we live in China or Japan) the Chinese characters. As adults we seldom stop to think about the mental-cum-physical process that turns our thoughts into systems on a piece of paper or on a video screen, or bytes of information in a computer disc. Few of us have any clear recollection of how we learnt to write.

Without writing there would be no history.

Unlike the writing systems in other civilizations, Chinese writing remains as a pictographic system (Fazzioli, 1986). Other writing systems have evolved into symbolic and then alphabetical systems. There are two sources for research about the origin and evolution of the Chinese writing system. One is legend as described in ancient Chinese books and the other is archaeological findings, such as pottery inscriptions, oracle-bone inscriptions and bronze vessels inscriptions. Legends in ancient Chinese books possibly lack credibility with respect to historical value. However, they do provide some positive suggestions about how the Chinese writing system originated. The main focus here is on the analysis of archaeological findings especially those on pottery inscriptions.

Inscriptions on pottery

Chinese writing began as pottery inscriptions which could be either pictographic or ideographic (Li, 1969; Chang, 1980; Murphey, 1996). Such symbols appear on Neolithic pottery dating back to at least 4,000 BC and painted with a hairbrush and knife. Some of these symbols appear to have been used for decorative purposes but others may have represented written characters including numerals, proper nouns, indicative words and logical combinations.

Oracle-bone inscriptions

Most of the early Chinese written documents now available are in the form of inscriptions on oracle bones and shells. They date predominantly from the late Slicing dynasty (1766-1122 BC).2 The main principles for composing Chinese characters had already been established in inscriptions on oracle bones and shells (Boltz, 1986; 1994). This suggests that the Chinese writing system, in the form of inscriptions on oracle bone and shells, had already undergone a long development and had progressed far beyond simple pictographs (Fairbank & Reischauer, 1989). Writing usually ran from top to bottom and from right to the left. This remained the rule until recent times. The more than two thousand characters found in the Shang exhibits are quite different to modern Chinese characters. However almost all have been clearly identified with later forms of writing. All the principles found in the eight or nine thousand characters commonly used in modern times were present in examples of Shang writing. The inscriptions on oracle bones and shells had been used by the people in the Shang dynasty to record political, economic and religious events. Because the people in the Shang dynasty were superstitious, the oracle bones and shells were used mainly for divining. Some counting records were found from the inscriptions on the oracle bone and shells, description below.

(a) No. 1 (fragment 16). Reading from bottom to top, there is "Dingxi time divined, King, fifteen dogs, twenty dogs, thirty dogs, fifty dogs; fifteen sheep, twenty sheep, thirty sheep, fifty sheep; fifteen pigs, twenty pigs, thirty pigs, fifty pigs".

(b) No.2 (fragment 17). Reading from top to bottom in the first right column, there is "Animal. Catching deer fifty six".

(c) No.3 (fragment 20). Reading from left to right and top to bottom, there is "Dingmu time divined, (missing words), good. Animal five (missing words) caught, catching (missing words), deer sixty two (missing words) one hundred and fourteen, pigs ten, rabbit one".

(d) No.4 (fragment 22). Reading from right to left and top to bottom, there is "Catching in place (missing words), catching deer eight, animal (missing words) one, pigs thirty two".

These records have been regarded as the earliest accounting records (Guo, 1986). No.1 is a record about expense and discloses numbers of animals used in divining. Nos 2, 3 and 4 are records of revenue indicating numbers of animals captured.

Inscriptions on the bronze vessels

During the Shang dynasty another form of Chinese writing also emerged which has been described as bronze vessel inscriptions. These Chinese writing inscriptions were generally short, consisting of only a few characters. The bronze inscriptions are different to oracle bone inscriptions (Boltz, 1986). The characters in bronze inscriptions are normally relatively naturalistic and "solid" in appearance, while those in the oracle texts are more simplified and abbreviated, written with thinner lines and dots.

Many of the Chinese writing characters in the bronze inscriptions are representations of birds and animals, fish, reptiles and other symbols in the prehistoric tradition. They appear either independently as personal names or family symbols or as part of the inscription denoting in addition the name of the owner or clan (Chu, 1973).

The technology of producing bronze vessels had been improved by the time of the Western Zhou dynasty ( 1 100-771 BC). The numbers of Chinese writing characters inscribed on the bronze vessels increased. In the latter period of the Western Zhou dynasty bronze scripts became more regular, with sharper angles and thinner lines (Norman, 1988). The longest of these numbered 497 characters cast on Mao Kong Ding, a well known Western Zhou bronze vessel (Chang, 1980).

Codification of Chinese writing in the Qin dynasty (221-206 BC)

During the Spring-Autumn and Warring States periods, many local variations of Chinese writing scripts were used in different States. Qin Shi Huang, the first imperial emperor, unified China.3 Subsequently the Chinese characters were codified and mandated throughout the empire (Keightly, 1989). These characters were based on the writing system used in the Qin State during the Warring States (476-221 BC) period and were standardised by the Prime Minister, Lishi. This writing system is called Small Seal Script and continues in use today in some circumstances such as writing posters and greeting cards.The development of Chinese writings in the Man dynasty (206 BC - 221 AD)

The Chinese writing system used in the Han dynasty became more simplified and a new system, called the clerical (official) script (lishu), replaced the small seal script as the official form of writing (Keightly, 1989). The Man dynasty is regarded as a significant period in the development of Chinese writing. Norman (1988, pp.65-6) writes:

The transition from the seal script to the clerical script and the subsequent universal adoption of the clerical script in the Han dynasty probably represents the most important transition in the entire history of Chinese writing. It marks the change from the ancient form of writing in which, despite a progressive tendency toward a more stylised and abstract representation, the essentially pictographic roots of the script could still be discerned, to a more purely conventionalised form of writing. At the same time another form of Chinese writing, cursive script, was also being developed. This was mainly used as an informal means of writing drafts and letters. It was a radically simplified system of writing in which strokes were freely joined together in order to obtain maximum speed and convenience. However, it became increasingly difficult to understand. Both clerical (official) and cursive scripts are now regarded as forms of Chinese calligraphy.

The development of Chinese writing after the Man dynasty

The standard (normal) form of the Chinese writing script, kaishu, which is still in use at the present time appeared during the Man dynasty and became the dominant form of Chinese writing during the period of the Northern and Southern dynasties (420-589 AD) (Norman, 1988). A new form of Chinese writing entitled running hand script (xingshu) also derived from the cursive script. This form of script adopted many features of the cursive script but kept its basic outlines much closer to standard script (kaishu). Therefore, running Chinese writing script might be described as a form of writing somewhere between standard script and cursive script but more popular and recognised before cursive script (Tang, 1965).

From the evolutionary elements of Chinese writing described above, it can be seen that Chinese characters became more simple and gradually lost their pictographic features. During the Man dynasty Chinese writing reached its peak. The forms invented at that time have been used for more than a thousand years. However, the process of simplification has never stopped (Murphey, 1996). In 1956, the Chinese government listed 515 simplified characters for official use. In 1964 more than 2,000 simplified characters were sanctioned. Many unofficial simplified characters have also been used by the general public.

Theory of the origin and evolution of Chinese writing systems

Two theories exist which are concerned with the evolution of Chinese writing systems. One theory argues that Chinese writing originated only from pictographs. This theory is popular among western scholars (Norman, 1988). The other argues that Chinese writing has dual origins, pictographs and ideographs. This theory was introduced by Xu Shen, a Chinese etymologist during the Han dynasty, and was subsequently developed and supported by numerous Chinese archaeologists. It maintains that pictographs and ideographs actually occur in the world simultaneously.
In line with Warbouton's pictographic theory, Boltz (1986, pp.424-6) believes that Chinese writing originated from pictography4 and evolved from a pictographic stage to a multivalent stage and then to a determinative stage.

Chinese writing originated, as all writing did, in what is commonly called pictography, that is, in the drawing of simple pictures to represent the word for the object in question. Each graph stood for a whole word, and the writing was therefore logographic, or in less common parlance, lexigraphic. In this respect Chinese writing was initially no different from all other writing. But where writing in Mesopotamia and Egypt, either because of the phonetic structure of the languages involved, or as a result of contacts with dissimilar languages, or both, moved away from a logographic script to become first syllabographic, and in some cases eventually alphabetic, Chinese remained essentially logographic throughout its formative stages.

The first stage of Chinese writing is called the pictographic stage. Each picture represented a word. For example, the character for "sun" originally was a circle with a line in it. The character for "moon" was still quite recognisable as a crescent moon. "Tree" was written as a sketch representing either the branches or the roots of a tree. The character for "elephant" was represented by a picture of an elephant. Some pictographs were quite complicated, such as the word for "sacrifice", which showed two hands holding a bird upside down over a symbol that meant "the spirits" (Fairband & Reischauer, 1989).

The second Chinese writing stage is the multivalent stage. In this stage, a single graph was invested with more than one value, either phonetic or semantic (but not both simultaneously). For example, the pictograph of an "elephant" was used to represent "image, or appearance" because they have the same pronunciation. Consequently both are described as homophonous words. Some pictographs were used as ideographs because they have different semantic meanings. That is, the pictograph of "mouth" was also used as the word for "name, or call". Similarly, the pictograph for "eye" was used as "see". However, while "the multivalent application of pictographs allowed the maximum use of the established inventory of pictographs", it "introduced the possibility of ambiguity" (Boltz, 1986).

The third stage of Chinese writing is the determinative stage. In this stage a determinative, especially a semantic determinative, is added to a graph to classify the meanings among the words. For example, a determinative "man" is added to an "elephant" to create a new word of "image". The semantic component is "man" and the phonetic component is "elephant". The word "image" is pronounced as "elephant".

Fairbank and Reischauer (1989) classify Chinese writing into three groups - pictographs, ideographs and phonetic derivatives. Pictographs, as discussed before, are words which resemble the meanings of pictures. Ideographs are those words which represent ideas rather than pictures. For example, the numbers, "one", "two" and "three" were represented by one stroke, two strokes and three strokes. "Above", was originally a shorter line above a longer line.
The third category of Chinese writing characters is phonetic derivatives. These phonetic derivatives are formed in two ways. One is called phonetic loan, which is when a pictograph is used to represent another word because they have an identical pronunciation. For example, the word for "wheat" represents the word "come" because they were pronounced identically in ancient times.

The other groups of words in phonetic derivatives are called phonetic, which are composed of phonetics and signifiers. The phonetics are "characters used to indicate the approximate sound of the compound character. The signifiers, usually placed to the left or above the phonetic, show the category of meaning to which the word belongs" (Fairbank & Reischauer, 1989). For example, the word for "branch" consists of a signifier of "tree" and a phonetic of "zhi" for its pronunciation.

Chinese writing researchers and teachers have been concerned with the history of writing. Many etymologists have written about the evolution and classification of Chinese characters. Among them, Xu Shen was seen to have been the first eminent etymologist in China (Norman, 1988). He wrote a dictionary, the Shouwen Jiezi, during the Han dynasty in order to analyse the structure of Chinese characters (Xu, 1963). Xu Shen's work remains a remarkable accomplishment. Today, the six principles (Liushu) utilised in his dictionary are still accepted as basic guidelines for graphic analysis of Chinese writing.

The main difference between the two theories about the evolution of Chinese writing is whether all writing systems originated from pictographs. Some scholars (Tang, 1965; Boltz, 1986; Keightley, 1989) support the view that writing systems begin with pictographs and then ideographs and other forms of words as later derived. However, many Chinese archaeologists (Guo, 1972; Cheng, 1973; Yu, 1973; Li, 1974; Ho, 1975) argue that ideographs were invented at the same time as pictographs and that they were not the derivatives of pictographs.

This argument is significant in terms of when the Chinese writing system was invented. So far, two kinds of drawings have been excavated; incised signs and pictures. The earliest incised signs were found in the Banpo site which has been dated between 4770-4290 BC, and resemble ideographs. The earliest incised pictures were found in the Dawenkou site which dates between 3600-3555 BC. This finding is inconsistent with the theory that Chinese writing had originated from pictorial drawings. Therefore, some archaeologists regard the incised signs found at the Banpo site as marks only, not writing characters, but no unanimity exists on the matter.

Why choose inscriptions on pottery to study the relationship between counting and writing in China?

The development of the writing system in China in ancient times had three stages: (1) pottery inscriptions; (2) oracle bone and shell inscriptions; and (3) bronze inscriptions. The archaeological excavations show that the writing system on the oracle bone and shells had been well developed by 1000 BC (Norman, 1988, p.58).
The Chinese writing appears as a fully developed writing system in the late Shang dynasty (fourteenth to eleventh centuries BC) ... . The script of this period is already a fully developed writing system, capable of recording the contemporary Chinese language in a complete and unambiguous manner. The maturity of this early Chinese writing has suggested to many scholars that it must have passed through a fairly long period of development before reaching this stage, but the few examples of writing which precede the fourteenth century BC are unfortunately too sparse to allow any sort of reconstruction of this development. On the basis of available evidence, however, it would not be unreasonable to assume that Chinese writing began sometime in the early Shang or even somewhat earlier in the late Xia dynasty or approximately in the seventeenth century BC.

The purpose of this preliminary examination has been to structure motivation and inspiration for inventing writing systems. The oracle bone inscription system had been used to record political and economic processes. It is not easy to understand from this form of writing system what were the initial motivations for the invention. Keightley (1989) suggests that the demand for writing was to assist people with problems of measurement and calculation. He says (p. 197)
I would suggest, therefore, as a hypothesis to be tested against future archaeological discoveries, that, on the basis of craft practice, social need, and actual graphic finds, the origins of Chinese wrung, ... are probably to be found in late-third-millennium (B.C.) sites of the Eastern Neolithic in China. He argues (that Chinese writing was more likely to develop in the region where social and craft complexity can be evident. However, this hypothesis has not yet been widely accepted. Based on the observation of pottery inscriptions in China, another hypothesis can be envisaged. This is that the demand for writing came from counting activities in line of explanations or justifications to provide authenticity. This has been associated with the origin of accounting (Littleton, 1933; 1953). The need to develop counting was one of the major tasks faced by people in the prehistoric environment. As productivity improved over time the need to keep accurate records of surplus became obvious. Several methods were invented to satisfy this need (Guo, 1984; 1988). The invention of knotting is a good example (Guo, 1988, p.7).6

In Early Antiquity, knotted cords were used to govern with. Later, our saints replaced them with written characters and tallies. In the ancient past, during the time of Rang Cheng, Xuan Yuan, Fu Xi, and Shen Nong, people tied knots to communicate. For a major matter, use string to tie a big knot; for a minor matter, tie a small knot. The number of knots corresponds to the number of matters to be dealt with.

The other method is the counting by cutting signs on wood or pottery, which is called writing inscriptions (Shuqie). There are many pottery inscriptions which have been excavated and they are the major resources for our research. Counting is a prerequisite for the development of accounting, as well as of mathematics, statistics, and other related disciplines (Guo, 1984). Gou (1984) discussed the difference between counting and accounting. Counting includes establishment of numerical systems (for example, the decimal system) and operational rules. However, for accounting purposes, it needs not only the numerical system, but also the establishment of measurement units. Therefore, "writing inscriptions for accounting purpose indicate three basic elements that cannot be omitted. They are the concept of numbers, counting rules and measurement units" (Guo, 1984, p. 18). Guo (1984) also envisaged that the first evidence of writing inscriptions for accounting purposes would have appeared in the patriarchal clan commune about 5,000 years ago.
Furthermore, the relation between counting and bookkeeping is captured by Howard ( 1932) when describing the emergence of science and accounting in the seventeenth century. The Ordinance of 1673 in France emphasised first and foremost the keeping of a journal "day by day all the transactions as rapidly as they occur" (p.92). These provide the numbers and prices from which running totals could be transferred to other journals and ledgers. These arithmetic processes beginning with counting have been fundamental in Europe. In the next section, the inscriptions on pottery from 4000 BC to 1200 BC will be studied in detail in order to test the hypothesis that people's motivation for the invention of writing was for counting related purposes. Pottery inscriptions have been found at more than twenty archaeological excavations (Chang, 1980). These sites date from about 4000 BC to 1200 BC. Only the most significant artefacts from these sites will be discussed.

The numeral signs among Banpo pottery inscriptions

The Banpo pottery inscriptions, which belong to the Yangshao culture (4770-4085 BC), were found in 1954. Among numerous pottery vessels and potsherds, 113 pieces bore 30 incised signs or symbols (Figure 4). The date of Banpo pottery inscriptions is about 4770-4290 BC according to the radiocarbon test. They are the earliest inscriptions so far discovered in China. There is a debate about these pottery inscriptions. Some believe that they are numerals and represent the origins of Chinese writing. Others disagree and suggest that these inscriptions are marks only. Keightley (1989) argued that it was unconvincing to believe that Banpo inscriptions represent the earliest attempt to create a writing system in China. He presented three arguments against this proposition. Firstly, the people of Neolithic China at that time had no need for a writing system because their culture was not yet sufficiently complex. Secondly, if the Chinese invented a writing system it would not have taken more than three thousand years to reach the stage represented by the oracle-bone inscriptions of the Sluing. Finally, Banpo inscriptions did not qualify as pictographs which are normally accepted as the beginnings of a writing system in other civilizations. Boltz (1986) held similar reservations. He said that there were three flaws in claiming that Banpo marks arc the earliest Chinese writing symbols: (1) Graphic similarity between the inscriptions on pottery and oracle-bone or shells was inconclusive, thus failing to provide evidence to show that these two inscriptions also have similarity in function or meaning. There is no such evidence available. (2) The Banpo inscriptions are not pictographs and this is contrary to the principle of the evolution of writing. (3) The time between the pottery inscriptions and oracle-bone inscriptions is over three and a half thousand years (from about 4800 BC to 1200 BC). It is impossible that "the nascent seeds of writing could have germinated in the mid-fifth century BC, but not grown into anything approaching a real writing system for more than three thousand years" (p.432). On the other hand some scholars believe that Banpo pottery inscriptions are the earliest Chinese writing symbols. Guo Moruo (1972), the paramount historian and archaeologist in China, believes that these inscriptions are undoubtedly symbols of the nature of writing when compared to those inscriptions on oracle-bone and on bronze at a later time. He even asserts that the simple ideograph occurred before the pictograph. Li (1974) believes that the Chinese writing system had already been invented before Banpo inscriptions. Based on what he deciphered from these inscriptions he concluded that Chinese writing had gone through three stages of pictograph, ideograph and phonetic development.

There are 30 incised signs that have been found on Banpo pottery. On the basis of the decipherment and interpretation by previous archaeologists, the inscriptions on Banpo pottery can be divided into three groups (Woon, 1987). The first group comprises numerals and includes five signs. Inscription "1" can be identified as the numeral "one". Inscription 2 can be identified as the numeral "two". Inscription 16 can be recognized as the numeral "five". Inscription 17 can be decoded as the numeral "seven". Inscription 24 can be identified as the numeral "eight". The second group has six signs which can be deciphered as pictographs or numerals. Inscriptions 20, 21, 22 were interpreted as a pictograph of steps on a hillside. However, these inscriptions were also interpreted as "forty" or "fourteen", "thirty" or "thirteen". Inscription 23 was interpreted as a pictograph of "a string of jades" or "fifty", or "fifteen". The final group has three signs which can be explained as pictographs. Inscriptions 3 and 4 were decoded as "an ancestral table". Inscription 12 can be interpreted as "a plant". Eighteen other inscriptions have yet to be deciphered.

Numeral signs among other Chinese writing pottery inscriptions found in China. The pottery signs represented by Yangshao culture (4770-4085 BC) have also been found at other places. One of these discoveries is of four pottery signs (Figure 5) in Lintong county, which date from 4671 -4545 BC. These are called Jiangzhai pottery inscriptions. The No. 1 sign can be interpreted as a numeral "five" or "seven" and the No. 2 sign is numeral "one" (see Figure 5) (Woon, 1987).

Chinese writing pottery inscriptions which include numerals have also been located for other cultural groups. Chengziya pottery inscriptions (2515-2340 BC) were discovered in 1930, near the town of Longshan and therefore named as Longshan culture. The Longshan culture (2515-2035 BC) is believed to be a direct descendant of Dawenkou culture (3605-3555 BC) (Woon, 1987).7 The Chengziya site had two strata. The upper layer was the Eastern Zhou (770-221 BC) ruin and the lower layer was connected with the Shang (1766-1122 BC) civilisation. Three pottery inscriptions were found at the lower stratum (see Figure 4). The No. 1 and No.2 signs are similar to the No.l sign found at the Banpo site. That was interpreted as "one" or "ten" by archaeologists.

The Erlitou pottery inscriptions (1625-1450 BC) were excavated near the town of Yanshi in north-western Henan province and belong to the middle of the Shang dynasty (1766-1122 BC). Twenty-four different incised pottery signs have been discovered here (see Figure 4). Seven signs may be interpreted as basic numerals. No.l sign is "one or ten"; No.2 sign is "two or twenty"; No.3 sign is "three or thirty"; No.8 sign is "five"; No. 13 sign is interpreted as "four"; No. 18 sign is "eight"; and No.23 sign is interpreted as "a combination of seven and ten".

The Erligang pottery Chinese writing inscriptions (1620-1595 BC) are located in the southwestern suburbs of Zhnagahou in Henan province. Thirty-seven signs (including variants) inscribed on pottery were discovered. Among them nine signs may be interpreted as basic numerals (see Figure 5). The details are as follows: No.l can be identified as "one" or "ten"; No.2 can be seen as "two" or "twenty"; No.3 can be interpreted as "three" or "thirty"; No.4, No.5 and No.6 can be deciphered as "five"; No.7 and No.8 can be identified as "seven"; and No.9 is believed to be "eight".

Wucheng pottery Chinese writing inscriptions (1530-1395 BC) were found in 1973 at the Qingjiang county in Jiangxi province (Tang, 1975) (see Figure 5). This discovery is very important to Chinese historical researchers because its site is remote from the Shang civilisation centre, the Yellow River Delta. The Wucheng site is on the south bank of the Yangzi River. However, the discovery at this site shows that these two cultures are strongly related. Sixty-six incised signs have been found and some of them can be interpreted as numerals. They include the numeral "five" represented by five signs (No.1, 2, 40, 41 and 42) and numeral "seven" (No.3).

The Xiaotun pottery Chinese writing inscriptions belong to about 1250 BC. Eighty-four fragments with signs have been found in Xiaotun site. They are the largest group of early pottery inscriptions so far discovered (see Figure 5). Nineteen signs may be interpreted as numerals. They are signs: No.l ("one" or "ten"), No.2 ("three"), No.4-8 ("five"), No.61 ("six"), No.9-15b ("seven"), and No.37 ("ten").

Analysis of numerals on Chinese writing pottery inscriptions

The Chinese writing inscriptions on potteries have not yet been fully deciphered. These inscriptions were scattered over many regions of the country and they have never been written in a single text. These problems result in difficulty for studying pottery inscriptions. Nevertheless, some scholars have made great progress in this area (Cheng, 1973; Li, 1984; Woon, 1987). In Banpo pottery inscriptions, there are five numerals out of thirty Chinese writing inscriptions, making up 17 per cent. Half of the Jiangzhang pottery inscriptions are numerals. There are three pottery fragments with signs from the Chengziya site, two of them being numerals. Seven out of 24 signs of Erlitou pottery are numerals. Of 37 pottery signs excavated at the Erligang site, nine of them arc numerals At the Wuchang site, there are 66 pottery signs, and six of them arc numerals. Finally, of 84 pottery signs discovered at Xlaolun site, nineteen are numerals. There are two exceptions. At the Dawenkou and Taixi sites no pottery signs with numerals have been discovered.8 However, it cannot be concluded that numerals had not been utilised by inhabitants at that time. There are only six pottery signs so far excavated from the Dawenkou sites and twelve from the Taixi site. The sample is too small to reject the hypothesis that numerals were not yet fully invented. Comparing the pottery signs from these two sites with pottery signs from other sites before and after that time, it has been shown that numerals had been understood and used by people at that time but simply had not been evidenced on these two sites. In general that numerals are among the earliest writing symbols invented by human beings. This indicates that one of the motivations of the invention of writing is the need for counting.

The frequency of occurrence of numerals has also been investigated. Li ( 1969) surveyed the frequency of occurrence of Chinese writing pottery inscriptions from some historic sites (Table 2) and this showed that the highest occurrence among the pottery inscriptions was of numerals. For example, in the Banpo pottery inscriptions the numeral "one" occurred sixty five times. The numeral "two" occurred four times, the numeral "five" occurred four times. In the Chengziya lower stratum inscriptions, the numeral "one" occurred twice. The message delivered from the findings, that numerals had the highest occurrence among the pre-historical inscriptions, is that one of the motivations for people inventing writing was for counting. People in ancient times needed to count their prey, tools and domesticated animals, so they invented numerals. Therefore, one of the incentives for the invention of writing systems, is related to counting, a first stage towards accounting justifications.

Conclusion

In this paper, the origin and evolution of Chinese writing systems have been examined. This showed that Chinese characters are different to those in alphabetical writing systems, and remain pictographic in nature. Secondly, the theory of evolution of a writing system was discussed as was its structure from two perspectives - the Western theory and the Chinese view. It was argued that the often accepted theory about writing characters being derived only from pictographs is not always applicable to Chinese writing. Chinese characters have dual origins: one is pictographic and the other is ideographic, especially with respect to abstract counting.

The main task of this paper has been to examine the relationship between early Chinese writing and counting. The earliest Chinese writing characters on pottery have been classified carefully. Although the evidence from this research is not as exciting as that found by Professor Schmandt-Besserat ( 1992a, 1992b, and 1996), it does show that numerals were among the earliest Chinese characters to have been invented. It seems probable that one of the motivations of innovation for Chinese characters was counting and, in terms of interpersonal rights and obligations as civility and laws developed among individuals and groups, of accounting. Under "historical evaluation" (Littleton, 1953, pp.85-91), accounting employs counting in conjunction with financial justification. This process needs to be explained by writing in books of original entries and by other environmental notes about numbers contained in the accounts. Given these relationships and conjecture about the emergence of the first example of an accounting system in China, the origins of writing and counting have been examined. When these essential elements of justification are developed then it may be assumed that conditions for the achievement of this general goal of accounting are in evolutionary mode.

Notes

1. In this paper, the authors do not intend to decipher any Chinese writing inscriptions. The roles of decipherment and interpretation of ancient Chinese inscriptions belong to the archaeologists. The purpose here is to discuss the linkage between these early inscriptions and accounting.

2. Reliable early inscriptions on tortoise shells were found in 1899. For many years before 1899, traditional Chinese medicine shops in Beijing had sold "dragon bones". These "dragon bones" were in fact old turtle shells and ox scapulae found in a village near the town of Anyang in northern Henan province. Signs were frequently found scratched on the surface of these objects. Two Chinese scholars in Beijing studied the signs and recognised that some of the signs were similar to the characters on early bronze inscriptions. Subsequent excavations and decipherment during the twentieth century have shown that the inscriptions are early examples of Chinese writing. They are records of divinations by the twelve kings of the later Shang dynasty who ruled from about 1400 to about 1200 BC (Chang, 1980).

3. Qin Shi Huang established the first imperial dynasty, the Qin dynasty (221-206 BC).

4. Boltz (1994) and Boodberg (1937) argue that pictographs are not writings because they lack phonetic components.

5. All Chinese characters are classified by Xu Shen into six categories which he called "the six principles of writing" (Liushu). The first two categories, Zhishi and Xiangxing, refer to words which are made of simple graphs and are non-phonetic in nature. The next two categories are Huiyi and Xingsheng and the overwhelming majority of Chinese characters belong to these two. The words under Huiyi category are non-phonetic. A huiyi ("joined meanings") character generally has two graphic components whose meanings taken together suggests another word. A Xingsheng character consists of two elements, one of which gives a clue to the semantic category of the word represented and the other a clue to its sound. The last two categories are Jiajie (loan characters) and Zhuanzhu. Jiajie are graphs originally devised to write one word, which have later been borrowed to represent the sound of another, often totally unrelated word. Zhuanzhu are from deflections, inversions or significant rotations of other words. For example, when "child" is written upsidedown it represents "childbirth" (Xu, 1963).
6. In China, an ethnic group called Miao in Yunnan province still used ropes to record their activities in the 1950s. For example, a man wanted to borrow two bowls of rice. He had to bring a rope to the lender's home. The lender tied two knots in his rope in the front of him. He left the tied rope at lender's home and took the rice. When he returned two bowls of rice, he would get the rope back (Wu, 1991, p.454).

7. A total of six signs or symbols have been found at Dawenkou, however, no numerals are among them. Tang Lan (1978), the foremost palaeographer in China, believes that the discovery of pottery inscriptions at Dawenkou is most significant to Chinese writing because these inscriptions resemble the features of the Chinese characters in the Shang and Zhou dynasties.

8. There are twelve pottery signs that were found at the village of Taixi in Hebei province in 1974. Among these signs no numerals were found.

References

Amiet, P., (1966), II y a 5000 ans Lex Elamites Inventaient l'ecriture, Archeologia 12, pp.20-2.

Boltz, W.G., (1986), "Early Chinese Writing", World Archaeology, Vol.17, No.3, pp.421-36.

Boltz, W.G., (1994), The Origin and Early Development of the Chinese Writing System, American Oriental Society.

Boodberg, P.A., (1937), "Some Proleptical Remarks on the Evolution of Archaic Chinese", Harvard Journal of Asiatic Studies, Vol.2, No.2, pp.329-72.

Chang, K.C., (1980), Shang Civilisation, New Haven: Yale Press.

Cheng, T.K., (1973), "The Historical Development of Chinese Writing", (Chinese), Journal of the Chinese University of Hong Kong, No. 1, pp.41-58.

Chu, Y.K., (1973), "The Chinese Language", in Meskill, John T. (ed.), An Introduction to Chinese Civilisation, Eexington, Mass., Heath, pp.587-615.

Fairbank, J. and Reischauer, E., (1989), China: Tradition & Transformation, Sydney: Allen & Unwin.

Falkenstein, A., (1964), Kelschriftforschung Und Die Alte Geschichte Vorderasiens, Leiden: E.J. Brill.

Fazzioli, E., (1986), Understanding Chinese Characters, London: William Collins Sons & Co..

Gaur, A., (1984), A History of Writing, London: The British Library.

Guo, D.Y., (1984), "The Historical Contributions of Chinese Accounting", Collected Papers of the Fifth World Congress of Accounting Historians, The University of Sydney, K4.

Guo, D.Y., (1986), The History of Accounting Development (Chinese), Beijing: Chinese Broadcasting and Television University Press.

Guo, D.Y., (1982, 1988), The History of Chinese Accounting (Chinese), Vol.1 and 2, Beijing: Chinese Financial and Economics Press.

Guo, M.R., (1972), "The Development of Ancient Chinese Characters" (Chinese), Archaeology, No.3, pp.2-13.

Ho, P.T., (1975), The Cradle of the East, Hong Kong: The Chinese University of Hong Kong and the University of Chicago Press.

Howard, S.E., (1932), "Public Rules for Private Accounting in France, 1673 and 1807", The Accounting Review, Vol.VII, No.2, June, pp.91-102.

Keightley, D.N., (1989), "The Origins of Writing in China: Scripts and Cultural Contexts", in Wayne, M.S. (ed.), The Origins of Writing, University of Nebraska Press.
Li, X.D., (1969), "Prediction of the Origin of Chinese Writing from Pre-historical and Early-Historical Pottery Inscriptions", Works on Origin and Evolution of Chinese Characters, Taibei: Lianjing Publishing.

Li, X.D., (1974), "The History of Chinese Writing and Its Evolution" (Chinese), The Quarterly Journal, Vol.45, No.2, pp.343-95, Research Institute of Chinese History and Writing, Taibei.

Li, X.D., (1984), A General Introduction to the Origin and Development of the Chinese Script (Chinese), Taibei: Lianjing Publishing.

Littleton, A.C., (1933), Accounting Evolution to 1900, New York: Russel & Russel.

Littleton, A.C., (1953), Structure of Accounting Theory, Illinois: American Accounting Association.

Mattessich, R., (1987), "Prehistoric Accounting and the Problem of Representation: on Recent Archaeological Evidence of the Middle East from 8000 BC to 3000 BC", The Accounting Historians Journal, Vol. 14, No.2, pp.71-91.

Mattessich, R., (1991), "Counting, Accounting, and the Input-output Principle - Recent Archaeological Evidence Revising our View on the Evolution of Early Record Keeping", in The Cost Heritage - Studies in Honour of S.Paul Garner, in. O. Finley Graves (ed.), Harrisonburg, Virginia: Academy of Accounting Historians, pp.25-49.

Advertisement

Mattessich, R., (1994), "Archaeology of Accounting and Schmandt-Besserat's Contribution", Accounting, Business and Financial History, Vol.4, No. 1, pp.5-28.

Murphey, R., (1996), East Asia: A New History, New York: Addison Wesley Longman.

Norman, J., (1988), Chinese, New York: Cambridge University Press.

Oppenheim, L.A., (1959), "On an Operational Device in Mesopotamian Bureaucracy", Journal of Near Eastern Studies, Vol.18, No.2, pp. 121-8.

Robinson, A., (1995), The Story of Writing, London: Thames and Hudson.

Schmandt-Besserat, D., (1992a), Before Writing, Vol.1, From Counting to Cuneiform, Austin: University of Texas Press.

Schmandt-Besserat, D., (1992b), Before Writing, Vol.2, A Catalogue of Near Eastern Token, Austin: University of Texas Press.

Schmandt-Besserat, D., (1996), How Writing Came About, Austin: University of Texas.

Tang, L., (1965), Chinese Characters (Chinese), Hong Kong: Taiping Publishing Bureau.

Tang, L., (1975), "Discussion on Wucheng's Culture and its Pottery Inscriptions", (Chinese), Antiques, No.230, pp.72-6.

Tang, L., (1978), "Discussion on the Nature of Dawenkou's Culture and its Pottery Inscriptions" (Chinese), Guangming Daily, 23 February.

Warburton, W., (1738), Divine Legation of Moses, London: printed for FIetcher Gyles.

Woon, W.L., (1987), Ch inese Writing: Its Origin and Evolution, Macau: University of East Asia.

Wu, J.N., (1991), Yi Jing, translated by Jing-Nuan, Wu, Washington.

Xu, S., (1963), Shuowen Jiezi (Explanations of Simple Graphs and Analysis of Composite Graphs), Beijing: China Publishing Bureau.

Yu, X.W., (1973), "Some Issues on the Research of Ancient Chinese Writing", (Chinese), Antiques, No.201, pp.32-5.

Wei Lu

Monash University

Max Aiken

RMIT University

Acknowledgements: The authors gratefully acknowledge the financial support of Monash University. They also thank the two anonymous referees for their valuable and constructive comments. The authors also acknowledge Thames and Hudson for permission to reproduce the material shown in Figure I and also acknowledge Harper Collins Publishers for permission to reproduce the material displayed in Figure 3.
Address for correspondence:

Wei Lu

Senior Lecturer

Department of Accounting and Finance

Faculty of Business and Economics

Monash University, Caulfield Campus

P.O. Box 197

Caulfield East Victoria 3145

Australia

Telephone: +61 3 9903 2972

Facsimile: +61 3 99032422