 The early Chinese writings on bamboo and wooden slips relevant to old Chinese phonology date from the late 5th century to the 3rd, mid 3rd century BC. Our interest in the warring states writings as relatively new sources of data for the old Chinese phonology mainly concerns the component structures of characters in the early Chinese script which are sometimes different from the received standard. So for instance in the Qin orthography for the word man, man meaning here well-known ask, you have the semantic ear and the phonetic door and in true and elsewhere the character has ear and the man as the phonetic. Here are some more examples, famous examples. The pre-imperial Qin regional orthography which is basically the direct ancestor of the modern standard orthography is represented by two different bodies of materials. One is the Qin manuscripts from the warring states period and the other is the Qin dynasties a small seal script as recorded in the shoumen jiezi 100 CE. The two are in different calligraphic styles the Qin manuscripts is in what is called the old clerical script which is quite similar to Han dynasty's clerical script and the Qin dynasty seal script is closer in style to the mid-warring states period true manuscripts than to the Qin and Han clerical scripts. So the seal script used during the Qin dynasty was a traditional and somewhat archaicized calligraphic style that was normally reserved for inscription but using orthographic variants as new facts about all Chinese phonology is complicated by many issues. One of the issues is that currently there are quite many different proposals of all Chinese reconstructions which significantly affect the way those previously unknown characters are interpreted and since almost all issues raised since Li Fangwei are still being debated we can say that there we have some five six competing reconstruction systems. There are many points that researchers of all Chinese today generally agree on for instance there's a distinction between labial villa and villa post villa initials we have a set of voiceless nasals and there are two liquids each of them had a voiceless counterpart medial takes care of the connection between division one and two rhymes of middle Chinese and the villa and post villa initials are closely connected. Xiaorong Fan in 1991 established that the voiced villa or laritual fricative or two origins one of them goes back with a guh and the other one goes back with the palatal glide so two origins one villa stop and the other maybe a uvular stop and all Chinese had consonant clusters yes we're sure of this fact just this much but what are the forms and functions of the clusters and we agree on the morphological function of the middle Chinese Qitong at the old Chinese stage we agree on the voicing distinction of middle Chinese is a morphological function at the old Chinese stage. On the other hand researchers still debate on many issues that make the phonetic shapes of reconstructed old Chinese words very different across systems. Many of these issues involve methodological assumptions that are unlikely to be solved through newly discovered materials unfortunately. So these are some of the issues whether to reconstruct voiceless stop initials endings or voiced stop coders for the middle Chinese open syllables which have all sorts of context with the middle Chinese voiceless stop endings so-called in root doi dran and whether or not to reconstruct a rounded nuclear vowel for subgroups of traditional UN and one and girl rhymes and what is the phonetic distinction between type A and type B syllables what are the forms and functions of consonant clusters should we reconstruct increasing s or valency decreasing n and how exactly to reconstruct the code are for those in young cases and what is the conditioning factor or factors for pelletization of old Chinese villa and post villa initials and what is the phonetic distinction between shown your third and fourth division words in old Chinese as well as middle Chinese and so there theoretical issues that we cannot quite solve through discovered materials and also researchers disagree on which graphic variants make meaningful new information for all Chinese and precisely what kind of meaning to make out of those character forms when they do not agree on how all Chinese words should be reconstructed in the first place so here is an example that everybody talks about the two script and elsewhere has this this phonetic man for one and the received Chinese standard it's also confirmed in Chin manuscript the phonetic is the door man and one interpretation is that man is older and was replaced by man moon because of the loss the neutralization of the rounded bow so this is a piece of evidence that supports the rounded bow hypothesis another interpretation is that when you don't necessarily accept the the rounded bow hypothesis these two are perfectly interchangeable functionally equal phonetics just looking at the archaeological findings it's hard to confirm actually the moon is older moon appears in many different regions whereas man appears only in Chin region but they're more or less contemporaneous previously the moon wasn't written like moon exactly it was a different graph one side was an icon I'm sorry to call it icon because it that graph did not survive in the writing system so we don't know the function we don't know the the meaning but but it's quite graphic it's a man listening obviously and there's the other component that is clearly this ear and here and in later warring states period meet warring states period we start to see this form one with a different phonetic this character went through a kind of irregular historical development which Liu Zhang described as bien-sing in a graph changes the shape to become a phonetic a graphic component in a character gradually changes shape to eventually overlap with a distinct graph that is effective as a phonetic component for what that the character stands for so if the change from the earlier form moon took place in the fifth century or later it means that the voiceless moon still existed at that time so the these characters is important for the date of voiceless nasal hmm not for the rhyme part that may be another interpretation it's just to make the point that you can make all meanings out of this one graph depending on what you think about all Chinese in the first place here's another example who the the word for skin who has the apparent phonetic who back initial but the old Chinese or middle Chinese has a labial initial and in manuscripts we have a graphic there is that represents the labial initial and how are we going to interpret this case one interpretation might be that it's the two sound changes into different dialects the K of back initial remains as back initial in the mainstream old Chinese but it can change to P in another dialect and that the P variant got borrowed into the mainstream old Chinese and that became the standard and another interpretation is the prefix P attached to K actually I didn't see in Bexar saga that they used this prefix to interpret the skin but this may be a possibility if you already have PK and KP alternating as clusters differences of interpretation influenced by differences in competing old Chinese systems are further complicated by the fact that researchers have introduced some new methodological assumptions to apply to manuscripts and not to transmitted sources and one of the assumptions is that the orthography of the manuscripts from the true region of the warring states period is based on the true regional dialect of that period but apparently the show on kids not supposed to represent the team dialect of the team dynasty another new assumption is that the phonetic elements in the characters in the pre-chin script represent syllable types rather than the pronunciation or syllables of particular words of course every graph has a syllable value but what's significant in this new assumption is that all graphs that have the same syllable type value are not just interchangeable but that they should be found interchanging for the same word this view suggests that there was such a fundamental change in the way that Chinese words were written down between preaching period to post-unit chin dynasty period in my view the change from the warring states to chin dynasty was not in how free the phonetic selection was but who was the authority to decide which forms are correct for which words having explained some complications about using manuscripts as new data I would like to explain some principles that I have in mind when I interpret the characters there are three points to discuss the first concerns the phonological periods and dialects represented by manuscripts manuscripts and orthography in them the second is on the principle of phonetic representation in the Chinese writing system and the third is on the vowel system of all Chinese by the first two I do not intend to suggest any new principle at all on the contrary I would like to reconfirm some some assumptions that sustained the many old Chinese reconstruction systems proposed so far first in any given body of excavated early Chinese writings we can reasonably expect the following four layers some four layers of phonological systems one layer is the mainstream old Chinese and the second layer is long words borrowed into mainstream old Chinese these words pronunciations from different phonological system these are words pronunciations from different systems phonological systems but once they are borrowed into the mainstream old Chinese they are at the lexical level not the phonological level so they may be used as synonyms along with their cognates in old Chinese or they may displace the original old Chinese words mainstream old Chinese words and the third and fourth layers are the standard phonology and dialects of the time when the manuscripts were copied the Chinese writing system is a logographic writing system is not a syllabary a phonetic graphic component in a character has a phonetic value it always does but it is it must be associated with a particular word or word family a phonetic is read once it is chosen for a character and therefore for a word it is read as a pronunciation of certain words that which it is chosen to represent and it's not just syllables the number of distinct graphs in the logographic writing system is comparable with the number of words in the lexicon rather than kinds of phonological units so what does a word mean in Chinese word means a morphine a morphine may be a root syllable with or without loosely attached prefix or half a syllable and yeah individual words as lexical items do not always correspond to separate etymons the identities of lexical items are indicated by differentiations of character forms and such relations between words and characters change over time and place in modern standard Chinese the word ya as in one ya is distinguished from ya as in yachi but they may be one and the same word from any more point of view carlin said that and we do change we have these three lexical items to support to branch and to limbs and they are homophonous chair from a ideological point of view they may be one and the same word and there's another item that is written as do many plus two and the meaning is many and it has a second reading as check this that's falling tone it's a chuteau derivative which is not distinguished graphically from their morphological variants so in a hypothetical warring states fruit this is a warring state style so yaa is the root this graph they represent the root and you add grass and you get sprout the character proper character for sprout and if the semantic is not specified it means that back tooth the true branch you add tree and it means branch that proper character for branch that word branch and it may be a graphic variant of two plus flesh which means limbs it's other also a graphic variants of variant of two or many and so on so the selection of a phonetic is phonetic component is arbitrary in the Chinese writing system a second and the third selection may occur in different regions without read or without phonological body we get motivation each phonetic component is arbitrarily chosen but once it's chosen it has to be a property of that particular word and word family and shouldn't be altered so readily so there's this word and and the middle Chinese initial is closed up or zero the zero is normally reconstructed as Q or it doesn't matter just it's reconstructed some sort of post-filler initial and but if it just happens that this phonetic series and contains only global stuff nothing else should we suspect that it made its own phonemic class Q and Q and Q and Q can interchange but the Chinese logographic writing system doesn't require these three separate phonemes to actually interchange and is it important that an is used to transcribe Alexander in the shinsi the yeah so the transcription sources are not as important as the more general principle mentioned just now but we incidentally know that the old sign of Korean transcripts and uses this word and to describe K I mean to transcribe K so the Q is more likely than the global stuff but this counter example is not important as long as we start with the assumption that the Chinese writing system is a logographic system I prefer to reconstruct to WA for those cases suspected as rounded vowels because it is difficult to establish the distinction between Kwon and Kwan we do not I do not deny the fact that TWV T TW vowel and dental ending and T T vowel and dental ending is a kai-ko-ho-ko distinction is strictly strictly observed in shesong and and rhyme but the problem is that we cannot tell the difference between Kwon and Kwan through shijin right we cannot use shesong evidence or word family or cognate relations because K and T will not make connections in those areas anyway so the only possible evidence is the shijin rhymes and we cannot confirm that the such distinction existed so why do we not see the distinction in the shijin we may have these explanations just ambiguous we don't have enough data and the change from to WA seems to have occurred quite early in some parts of the shijin and the text was edited in hand times and their irregular line arrives and so on but what if the distinction never existed so this may be an alternative interpretation Kwon sounds like Kwan and it's phonemically Kwan because Koo is natural so we get the initial Koo and An and Kwan will contrast with Kan in shesong series or word family relations on the other hand An sounds like Kwon and Kwon and phonemically it doesn't matter whether it's analyzed as Kwon or Kwan because it won't rhyme with Tan anyway so it's interesting to compare the W with medial R the old Chinese dental initials followed by medial R becomes vitro flex initials in middle Chinese in the case of KR the vitro collection does not occur because Kerr is not natural so when the Kerr doesn't occur occur that R may affect the vocal part more significantly so that's how in some modern dialects the Chongyue third division vowel turns out as rounded whereas the same effect the same result did not happen with dental initials so there are many complications in using manuscripts but once we have an old Chinese reconstruction system and principles that are consistent with the ones underlying the system then it's possible to sort out those phonological layers and manuscript materials on the basis of what we already know about old Chinese going to suggest that these two phonetics originally stood for the same syllable value and the value was like prior with the R code the first graph is identified with the standard button and there's no significant orthographic difference between this true graph and the receipt and is synonymous with pay also meaning riverbank slope and setting pay and ban are synonyms and in true manuscripts also in the seal script the two phonetics look quite similar but originally the B didn't look like originally it had two components as carbon identified with a tide enhanced components that didn't look like so what happened was that because they they are created separately to stand for the same syllable and of course they would they would write this members of the same word family so they came to eval evolved to look more and more like each other so that's what we got in the warring state script so with this application we can revise the old Chinese reconstruction forms of these words oh yes Dexter and Saga identified the change from art to art as a feature of a Shandong Chida and Chida had some prestige in the ancient Chinese world versus line suppose there is a high a bit of true wishes it's time to speak the Chi language should be higher native of Chi to tutor him or native of true to tutor him so in this position we revise the reconstruction like this so at stage one they were cognates and at stage two they were synonyms because one was borrowed into the the other variety and Dexter and Saga has this word button perverse and we have many words with the other phonetic that has similar meanings basically slanting partial one-sided insincere oblique slope walk lane lean to one side slanting of the partial as another example there are two phonetics alternating for the same word deep round ta and ye were not distinguished in the early script they were one and the same graph and so there are the two variants not three actually in the early script for this word and yeah the the other one the first one is true and chin scripts and the other one one is preserved in Jin script so what this means is that D the word for D ground has R ending and then it evolved to J ending I mean open ending so in this case the Chi dialect variety got the borrowed into the mainstream and then it replaced the normal one the mainstream one which would be an ending so in these two examples we identify two kinds of dialect variation but don't be insertion or deletion or rounding or unrounding if you use the rounded well I help us is we don't lose much if you choose either either high policies and I'm and I came from earlier on the dialect borrowings in mainstream all Chinese increased synonyms in the lexicon and influence graphic evolution thank you very much