 It's a great pleasure to welcome Professor Nathan Delphi, who is a professor of Chinese, a Chinese language, having traditionally been a terrorist for a number of years. Now, I don't think at this time and now we've had a talk on even if they've known Chinese before. Well, I don't know if I remember, but Nathan is a very well known historical so we're very pleased that you could talk to us about that. I would say on many things. So the paper has probably two halves and you will notice because the first half I will be doing additive, but then the second half becomes a bit more explicitly political. I will actually read my text. So first, I'm going to avoid Chinese altogether and look at correspondences between Tibetan and Burmese that motivate the reconstruction of uvulars in whatever their common ancestor was, which maybe we'll call Sino-Tibetan for now. So there are two correspondences between Tibetan and Burmese. One we have a velar in both languages and one we have a velar in Tibetan and we have a zero initial probably articulated with a lot of stuff in old Burmese. And I think that other things being equal, the typical way you would address the corresponds like this is you would reconstruct the velars that are velars like in both languages to a velar and you would reconstruct the velar zero alternation to something else. And how about a uvlar because it's quite normal for uvlers to turn into velars. It's also quite normal for uvlers to turn into long stops. So now we reconstructed a velar uvular opposition for the common ancestor of Tibetan Burmese. But let's look at some data. Okay, so we have the word for bitter with both velars in both languages. We have burden or load in Tibetan which I'll just say none of these cognitive proposals come from me. They're kind of widespread in the literature. So that's compared to saddle frame in Burmese. And then we have kidney and or I don't actually use this word so I'm not quite sure how to use it. It rains. And then loin or waist in Burmese opposed contradict in Tibetan and block or obstruct in Burmese and then hill in both languages and the first person singular pronoun in both languages. Now looking at this you may have some questions about details which I'm basically going to not go into. But I will just say the main phonemic opposition in Tibetan is voiced versus voiceless. So you can just ignore the aspiration. And in Burmese on the other hand the main phonemic opposition between non aspirated and aspirated and there are no voiced consonants in inherited Burmese vocabulary. So that simplifies the picture. But then you'll still say okay but in the word for hill for instance Tibetan has a voiced initial and Burmese has an aspirate which will probably go back for voiceless. So there's still some work to be done there. And then you can start telling stories like well the aspiration or the voiceless is conditioned by the s prefix or but how do you but it's a mess. One way or another manner or responses even signers that are messy and there's a big chunk of my 2019 book that lays out that mess. And it's messy because so much of the morphology in probably the further language and certainly in many of the other languages in prefixal. So if you have a lot of prefixal or follow you you'll get messy initial manner of correspondence. There's a lot of work there to be done and you know if you some weekend are not sure how to spend your time. Okay so so that was it for velars and you know let's say and it's pretty simple. Got velars in both languages. Okay so now let's look at the uvular examples. We've got needle in in both languages home in both languages and I'll say I particularly like this document because you know needle could be probably is in fact maybe a loan word but home is probably home. Yeah and then we have to fasten or suffocate a card with just squeeze and I'll just point out that this rhyme and goes back to it that's a normal sound change in Burmese so it looks it is better than it looks. This comparison of blood and ashamed is actually one of my proposals and let's say the only person who's taken notice of it is Guillaume Jacques and he projects it but it works in the phonology and like I think it works in the phonology and I think it's very easy to imagine read having you know I think the semantics work and the phonology works so the reason that Guillaume rejects it is because he wants to do other things with the Burmese word. Yeah and then this space or opening so again you just say that this is so it looks like there was a thing that was turned into a velar in Teven and is just lost in Burmese and luckily because it's just lost in Burmese this question of the manner of correspondences doesn't arise. Okay so so you know so far so good we've got velars and we've got uvelars in signed Teven. Some people in the field would say yeah but it's very artificial if you chose two other languages to start with you would maybe end up with different results and I think that's true but so what like we need to start from somewhere and starting with you know two old literary languages is a normal thing to do. They are not particularly closely related. Now let's add Chinese into the picture and here I'm just sort of mechanically saying like well I've already given you the cognate sets like or can we put Chinese edamame into those sets. So we generally can't yeah so first let's go through the velars yeah so we add Chinese word for bitter we add a Chinese word for carry and then you say what's his age with the little thing under it that is a symbol I'm using my system for writing middle attorneys and if you want to know about it you can read an article where I lay it out but this is a voice velar primitive and everyone agrees that one of my sources maybe it has multiple sources and he's a girl but just a voice velar solid yeah so you can look at it and think that he kind of looked up okay so then we have the kidneys and and then you know in Tibetan we add liver for Chinese we have this opposed construct and we add she'll reward off in Chinese we have hill and we have a word for hill we have the first person pronoun and we add the first person pronoun and I will just point out to this ooh this middle Chinese ooh comes from the rhyme ah and actually uh you know you know how uh like what's the sound of sheep making great right yeah so so in the same way uh in Chinese the way that you say that your surprise is ooh yeah so uh probably ah ha right okay so um yeah so that's that's something that did everyone also agrees on the ooh comes from ah and that rhyme is called fish rhyme in middle Chinese to terminology uh and for instance the word for fish in it in uh Mandarin Chinese uh and the Tibetan word for fish it means it's not okay so now it makes you the vowels work so I've added the Chinese data into the velar correspondence and lo and behold they're velars so now we can be you know more happy with having reconstructed the address for uh pro-design to them well now let's look at the uvelars so we add uh needle and we get chim okay so this has some problems we have a ma instead of a bug that's the big problem uh and then we have a shit where we want something other than a palo but in old Chinese the paloes have two sources and one is dentals and one is velars the velars parallelized very early and then later dentals parallelized um and uh with with velars the conditioning environment is a little bit uh insecure but it won't surprise you that the the velar is one of the uh places here I think we reconstruct usually a schwa so the Chinese would be something like um so it's not a great comparison and that's one reason why I think maybe uh it's a bond award and I I think although I really don't know about these things uh that other language families in the area like fengmen and kradai have similar words for needle but of course they put the bar in Chinese yeah damn needle the problem uh but I'm not going to put too much weight on it so if you don't like it throw it away uh okay then house and uh this one I really like this comparison comes from the Laurent cigar uh so so it means subterranean room and it refers to um it's actually I I read this this book about um records of the inquisition in France in the 14th century and then at the time the French lived in like in that part of France which the southeast they lived in this kind of house it's like half underground where you sort of dig a round hole and then you put a roof over it yeah so uh that's what this means specifically and you know I don't know what kind of house the photos are defense lived in but uh it seems like this is probably you know the maybe the inherited word for house in Chinese and that uh another word for house is this means you know mezzo net or something like that okay so uh so I think that's a really nice etymology and then you would see it like for me it has a zero initial okay uh and then uh for for squeeze you say that looks like a totally random word who've chosen uh what it can come from x in in old Chinese and that again it's unparalleled that that um basically what happens is uh the k s uh the the k drops out uh and then you get blah blah s yeah and then the s changes into a high ton yeah so uh right and then we have uh this word red or fiery uh that basically gives me exactly what I want for for my comparison in this in this last example and then I'll just say that this ah vowel so it's something like hat everyone agrees that one of its sources is so we can we can temporarily reconstruct something like uh no problem yeah this is this h I'm using was uh in middle chinese a voiceless be decorative and the origins of the voiceless be decorative is a serious problem for old Chinese but uh you know I feel like this fog is good enough that I'm allowed to keep it for now yeah and uh for now I would just point out that none of these reflexes are the same as the ones in the velar cases except maybe for needle but but but even let's say just at phase value even needle have a different outcome than uh the velars that he was looking at earlier right so that suggests to me that that we do need some kind of velar versus something else distinction uh in old chinese but even if you are happy to throw out the first and the last example then we still have two quite good responses and you know maybe some people would say you shouldn't build the phonology of a pro-language just on the basis of two examples and I would say what well then you and I is great the whole point is uh sound trade is exceptional so if it works and no that's my feeling I'm happy to build the case for uvulars entirely on the basis of these two examples uh I'm happy to say I don't think I need to do that but I but if I didn't need to do that I would be okay with that okay so I haven't so far made the case for reconstructing uvulars in the history of chinese uh on any chinese internal graphs yeah and that's uh you know that's the way I'm trying to sort of structure this this paper is to say look we need velars and uvulars before we get to chinese so then the question isn't you know whether or not there were uvulars in the history of chinese we know there were uvulars in history of chinese the question is what happened to the uvulars and when did they become something else we also know that uh they did become something else because there are uvulars in chinese today right so it's a question of when it's not a question of weather and now the question is did old chinese have uvulars well some people say that old chinese did have uvulars so now I'm going to present that chinese internal case for uvulars all right so to start with just some basics about how the chinese writing system works so some chinese characters are logograms like the word for horse is a picture of a horse and in an earlier version of this handout I made it very big so you could see it but it took up too much room because I made it small again but if you turn your handout sideways and switch you can say okay now they they didn't turn it sideways they wrote their horses like this like standing on their tails yeah so I mean I don't have a story to tell about that um okay so some words are like this word for horse is a picture of a horse and you read it as your word for horse uh and then most characters have two parts one the semantic indicator and one the phonetic indicator um so for instance uh we have this word headboard uh and it has a semantic tree and it has the phonetic horse so uh so that if you see this character you will think how am I supposed to read it well it's either a word that has something to do with horses and is pronounced like tree or it's a word that uh has something to do with wood and is pronounced by horse and then you say oh it must be headboard because headboard is pronounced the same way as horse yeah that's the way the script works uh at least you know like we're imagining the early days of the elaboration of the script and I'll just point out it's exactly the same as Sumerian yeah the way Sumerian is different than from Chinese is the Chinese came up with a clever idea of how always having the logogram fit inside a box yeah whereas in Sumerian you can kind of just jumble these phonetic and semantics around and get kind of long stuff I mean I don't even know what they call it though like like we would say you know Sanskrit context but that's not quite ready either yet um but uh you can have you know you have these little graphic components phonetics and semantic terminers and you can pile them up uh you can pile them up in Sumerian you can pile them up in Chinese in Chinese that stack always has to fit inside a box that was not true in Oracle loans but by the time you get to the joe diamond speed someone had this idea of oh we should have um the characters fall be the same size no matter how complex they are so for the purposes of historical technology that means we have a huge pool of information although how to use it is quite uh tricky where you know we know that horse and head board must have sounded somehow a lot maybe not the same right so what are the parameters of phonetic similarity that uh permit you to use this is called the repus principle right it's like it's like on uh that god there was some game show on tv when I was a kid uh that I don't remember the name of where where they would draw pictures and you would have to guess the sentence yeah and so they would like always use an eyeball for the word for the person yeah that's the repus principle so what is the scope of phonetic similarity uh that this script has in mind for application of the repus principle it's just important to sort of get out of the way that probably in reality different people at different times had different phonetic criteria for their application of the repus principle but if that's our understanding we will get nowhere I just say well you know some people probably did it differently at different times then we can't use this evidence for research so instead it's useful to articulate some hypothesis of what those parameters of phonetic similarity uh were and uh and I will get to that in just a moment but first I will just introduce the term shesham series the shesham series is a suite of chinese characters that share the same phonetic yeah so I give one here for you so um so you can look at it yeah so we have the kind of mother character which is the lobe graph so it has no semantic determiner it it's like your picture for worse it's it's just all symbols that stand for a word and then you have all of the characters that are sort of born from it and I have a way to transliterate this it's inspired by how things are done in the Near East where I write the for instance the mother character in this case I write it as a pa and then I continue to write it as a pa as an indication of the differentiation in all the other characters that's transcribing the phonetic determiner in those characters uh but then I use a sort of funcated latin uh abbreviation for the semantics yeah so the first one is vest pa that means it was pronounced like pa and it had something to do with clothing then uh I have tom pa that means it was pronounced something like pa and it had something to do with a mound yeah so uh if you know Chinese which some of you do you'll I think see what I'm trying to do here yeah and I think it would be really nice if other people did it this way because then people who don't know Chinese characters could make sense of arguments about Chinese is world penalty in the way that happens in the study of Sumerian and and Hittite and what not uh but uh let's say everyone who knows any Chinese who has seen this thinks I've absolutely lost my mind so anyhow uh and I have an article quite old one now where I lay out this sort of system and then you see you have two system two native systems next to each other the first one is to how I transliterate the character that's the point it's not a reconstruction it's a transliteration uh and then the second one is the middle Chinese pronunciation the middle Chinese pronunciation is attested okay it's not attested in roman letters we would need a lot of time to talk about what's going on here with with this romanization but it is attested there's a book from six or two that systematically represents the pronunciation of about nine thousand characters so you can understand this as that's how it's still a logically attested in 602 and and uh I have a transcription relying on bachelor transcription sort of I've cleaned it up to make it more indelogical because that's how my brain works uh and it represents the attested pronunciation from 602 so you have an attested pronunciation from 602 in italics and a transliteration of the structure of the character uh in non-italics okay so now you know what a shesham series is so any observations about uh how similar uh the pronunciation of a word has to be in order to warrant being written with the same phonetic and don't look at the handout for the answer anyone any just go or do a look at the handout but just the don't look at the answer to the bottom page yeah any any observations what do these have in common yeah and yeah and yeah place of articulation farm place of articulation place of articulation yes okay uh uh and uh now let's look at the I don't know the the the outslout do we have uh do we have uh big and you know tongue and whatnot in this series no yeah no so so there's so so the the initials have to have the same place of articulation and there's some kind of constraint operating on the on the coders uh you would say well they're not all the same rhyme but actually it's it's only two and um that probably reflects the latest split so probably in old Chinese they had the same uh rhyme uh and you'll notice that the I write an I before the E in half of them that's the kind of orthographic convention we don't exactly know how to pronounce so in old Chinese there were two types of syllables which are called type a and type b but in this case we have type a and type b so we can hypothesize that whatever the type a b distinction originally was in probably the kind of middle of the Han dynasty it conditions some kind of big vowel change so that's why we have this uh these two uh different rhymes and in general it's always two yeah you you you you have a traditional series it's long and short uh it might be only one it might be that that particular traditional series is all type a character uh but it it never has you know 10 different vowels so it's one or two so we can hypothesize that in old Chinese there's always one yeah okay now what doesn't it represent well even here we see that it doesn't we're indifferent as a tone yeah we can say uh the tone doesn't matter and the manner of articulation does matter one and the thing and that it was fun about the uh the the chapter with James was we talked about methodology historical and um back from cigar season solstice variants where you have a hypothesis and then you look for evidence to refute that hypothesis I don't think that this is a good philosophy of science uh and and I am going to promote the exact opposite so his is called the hypothetical deductive uh and I'm going to follow my empirical dogmatic though uh so the empirical is we look at something and we see a pattern yeah and then the dogmatic is I I'm committed to this pattern now yeah so I have to continue to see this pattern even when the evidence contradicts it yeah um and so that's yeah that's what I call the the um the empirical dogmatic method and I and I think all major successes in uh historical physics have come with this method think about the days of sword and the laryngeals you can think of it that way you know he knows the pattern and then he saw the pattern even when it seemed not to be there yeah so I'm going to now apply that uh method and in this case uh we'll call the shesham hypothesis where the shesham hypothesis is if two words are written with characters that share the same phonetic then they would have had the same place of articulation uh for their initials and the same rhyme for their finals and uh if I find a shesham series it appears to violate that uh that's my problem I need to fix it okay so now let's look at one such uh style of violation of the shesham hypothesis there are various phenomena which are basically the types of of violent apparent violations of the shesham hypothesis so now we're going to get one particular final phenomenon uh which I would uh you know in the vision of naming things after people I would call pawns phenomenon so in pawns phenomenon we have contact between the velars and we call them laryngeals in uh in Chinese studies and and don't put any particular stock in that joint word it's okay but you'll see that it's not an actual class the global stuff a ya the voiceless velar predictive yeah so so we have we have contact between velars and other stuff yeah okay so uh let's look at it so in the first so in example one now now just head straight to the example the body examples in the same series you see we have a velar a glottal stop a voiced uh velar predictive and a voiceless uh in example two we have uh an aspirin velar stop and a yup in example three we've got a glottal stop we've got an initial velar well it's with a glottal stop anyhow you get the idea we we we have things that are not homo organic in middle trends so uh in baxter and saguerre's theory there are the sound changes that you see immediately after the name of the example so for impen that a voiceless uvular becomes a glottal stop that an aspirin uvular uh becomes uh velar predictive that a uh voiced uh uvular stop becomes uh voiced uvular sorry voiced velar predictive but only in Taipei syllables so yeah you know i just commend you to the study of these sound changes uh for the Taipei and say these are the proposals that Baxter and Saguerre made to fix these broken shesham series and this is the dogmatic component we say okay they they violate the shesham hypothesis but if i believe this they don't anymore okay so so far so good we we preserved our shesham hypothesis yeah from this potential attack yeah not quite so easy because you notice that the the hypotheses are all about explaining the non-vealers as coming from uvulars but these series also have vealers in them yeah so if i have a series it has some uvulars and some vealers it's still a violation of the shesham hypothesis so now i also need a theory about how the vealers got in these series yeah so far sort of examples one two three four are about how the laryngeals come from uvulars now i also need some of the vealers to come from uvulars and Baxter and Saguerre's uh proposal which is quite mechanical but not inlawful ideologically is that there was a prefix that uh protected if you like the or or it it conditioned a bleeding relationship right so so uh if i had a cue it it sort of wanted to turn into a glow stop but if i have a prefix in front of it then it doesn't it changes and it will be like so it's perfectly reasonable although a little bit formalistic now in many cases for particular etymologies they do have a story about what the prefix was oh here it was a t here it was b but that involves all sorts of things that i don't want to touch on and is definitely one of the most controversial parts of their reconstruction system that they do sort of use as a skeleton key to solve all problems so here i'll just say you know well you know we we can do it by just saying yeah there was a conditioning environment that meant some of the uvular terminus the vealers now it's still let's say has a kind of real utility in terms of shesham series because there should be because let's say we have this hypothesis now that there are two kinds of vealers right in middle Chinese there are the vealers that come from vealers and they're there are vealers that come from uvula and now we would expect that those behave for instance in in in terms of what topics they have you know languages but also potentially you know if they were actually if this is true then you would expect it to be philological evidence also that they were different yeah and we'll get to that okay so now is when i switched to reading the paper so uh this is the first set of conclusions and the first half of the paper the reconstruction of a vealers versus uvular opposition may be positive purely on the basis of a correspondence between the vennemperities once this reconstruction is made the fate of uvulars in old chinese requires some kind of account the null hypothesis would be that signage event vealers persisted as vealers in old chinese and then likewise signage event uvular persisted as uvulars it may be the case that the evidence in favor of this null hypothesis is so far not overwhelming but this is not a particularly germane objection because it's the null hypothesis and the null hypothesis gets the benefit of the doubt in any event because uvulars are already required in signage benton on grounds unrelated to chinese one cannot dismiss the reconstruction of vealers within the history of chinese by brandaging often great you can't say oh don't get me you're yours you need the uvulars so now the the problem is that axl schussler doesn't buy this so now i'm going to look at what does axl schussler say and what do i have to say about it so uh if we look at the treatment of a few shesham series in the hands of schussler that is minimal chinese written ocm and in the hands of baxter cigar which is new old chinese and oc um it's pretty clear that baxter cigars reconstruction satisfied the shesham hypothesis better now you you know that's maybe axl schussler would say that's because they you know stop at nothing to but let's go to say at the starting point you know we all use the situation hypothesis and they stop at nothing and manage to to get there whereas he stops the first time he trips over a bevel in the road and he doesn't get there and uh this is in uh table one where i've broken it into what is it one two three four five different series and you can just look at baxter cigar reconstruction and truthfully reconstruction and i think that the second chunk is where it's most right where uh you know for ocm we have clone clone on long these things just basically have nothing to do with each other in terms of emission yeah whereas it's it's lovely if one eight uh or baxter cigars uh reconstruction okay so i would say free with fascia you know we should say good job baxter cigar you have stuck with the shesham hypothesis uh and shussler correctly characterizes the effectiveness of the euclid approach namely this uh this interference of laryngeal in uh velar series that point a point b the failure of certain middle chinese homophone groups to overlap in the writing of activated text is basically we expect certain kinds of spelling mistakes if two things are not the same you know like we could say for instance night uh with a k and night without a k uh we would expect at some point to get mixed up if in fact they were pronounced the same and and we have cases where it doesn't happen in particular these two things go and i don't have a good way to read them out loud but let's say one is maybe gong uncle all college and one is gong work so uh baxter cigar say they don't get confused so they were probably pronounced differently uh so schruppler gong is that and he also mentions that there's foreign transcription evidence in favor of the euclid hypothesis and i'll look at these three types of evidence but nonetheless schruppler rejects the euclid hypothesis so we'll look at uh these three types of evidence and what he says so for contact between healers and rentals which was the reason that i motivated the reconstruction of uh euclid in the history of chinese so schruppler makes an effort to defend his belief that phonetic series are an unreliable kind of evidence which can be overlooked and convenience and quoting baxter cigar there so now i'll just read this long quote from schruppler it says old chinese rhymes uh with many words and graphs that's to say you know i have a lot i stumble over that every time rhymes with i think we're talking about country that's not what's happening here right i think it's old chinese rhymes that have yeah many words and graphs like ah have ample graphic elements available heat words with initial dealers clusters and glottal stop separate so that k and glottal stop do not need to mix thus we have this series which writes only uh dealers and then we have jaw and jaw that write clusters like and then we have this this that writes well stop initial so in this case we've got lots of different series we can make very fine brain phonetic distinction on the other hand rhymes like up with relatively few words have correspondingly fewer graph themes at their disposal there is no phonemic sorry there's no phonetic series has no shen sheng series um up so the few words of this shape are slipped into other series once it was done occasional mixes of dealers and individuals were accepted so it's if i want to have a two-part store i have to do something for pragmatic reasons and then i allow myself to do it in other circumstances so bachelor and cigar agree with schruppler that some phonetic series were exceptionally pressed into service to write rare syllable types a phenomenon they they doubt both demure phonetics but they reject schruppler's intention that these uh full demure phonetics were precedent setting for more common simple types so this is what they say they say the preference for optimal phonetics is local in scope that a word pronounced as up could be written with a foot demure cup phonetic did not undermine the entire symptom i.e. did not result in a phonetic for example con becoming acceptable way of indicating pronunciation on when there were already phonetics uh for on uh that existed sorry i'm not reading the quote very well so schruppler's uh notion that any phonetic contact that can occur when one is hard pressed for an appropriate phonetic then gives rise to precedence for similar contacts in more common syllable types leads inexorably to a testable prediction such contact should start in rare syllables uh and then spread in the course of of younger character coinages yeah so some characters are a little better than are invented at a different time so these contact should spread through time for more and more common syllable types schruppler offers no discussion of the relative chronology of the character coinages uh in question uh so he doesn't you know so there's a there's a prediction and he doesn't uh say anything about whether uh the evidence is in his favor or not and back through cigar say it's not so this is what they said they say in fact throughout the old chinese period the match between the system of phonetics and the language of spinology did not fade as schruppler's model would predict it grew stronger to the point that towards the end of the period the system of phonetic had become a near syllabary warren states scribes writing on bamboo slips kept phonetic elements relatively stable while semantic determinatives could vary from one token to the next even under the same hand referring to the data in the first two rows of table one schruppler contends that shadow shade and tang right are not derived from the same root but an example of a popular mental connection between grass that has nothing to do with etymology the reader is left to wonder how schruppler has gained direct access to the inner lives of the ancients factor and cigar point out that in early texts both words are written with the same character fact that by itself towards schruppler's account since the rebus principle is only available when two words have similar enough variations so quietism is the only disposition to the contact of billers and laryngals in phonetic theory that schruppler will countenance as such in his discussion of the history of the proposal he scolds pawn and baxter for upturning his apple garden inconsistently he makes fun of pawn for restraint and then immediately mocks baxter and cigar for their wood spot baxter's openness to revising his thinking in the course of 22 years of the digital study schruppler also ridicules so let's see you can do these things pawn postulates uvular stops to explain co-affirmances with vealers stops although one may wonder how k q contact works better than glom stop k contact that was accepted until recently by scholars including baxter and has felt natural to native writers since pawn times at least but unlike pawn baxter and cigar required distinct vealers and uvular series which neutralizes pawn's original rationale for uvillars which was precisely the coincidence of the two mannered series vealers and uvillars in phonetic series this should have been the end of thought but theories have a way of getting divorced from their raison d'etre and taking on a life of their own baxter and cigar now use uvillars to claim all chinese distinctions among middle chinese vealers the transition from this is me now right the transition from pawns to baxter and cigar's proposal is not a story of a hypothesis losing touch with its raison d'etre but rather of a raison d'etre finding more adequate fulfillment in a revised hypothesis the raison d'etre of the uvular hypothesis is to bring the shesham series that makes uvillars and laryngals into conformity with the shesham hypothesis the formulation of pawn falls short of this but the formulation of baxter and cigar achieves it as baxter and cigar point out other major advances in our understanding of all chinese concepts have followed exactly the same reasoning and in those cases shusler pursues his quietism the mixing of nasals with voiceless voiceless stocks or precatives in a phonetic series with no legs or no more a violation of the shesham hypothesis than the mixing of k and blah blah blah i haven't given these examples but uh you have series that makes ta and na for instance uh but whereas shusler is content to eliminate the former by reconstructing voiceless nasals that's what most of the field does he vaults at reconstructing uvillars to explain the latter why not propose that it is a lack of appropriate phonetics for a syllable such as c that led the desperate ancients to take recourse to the character knee ear uh to write a word like p shame or disgrace uh before one turns to a non-synetic sound such as na and there i'm non-synetic is a term that uses for uvillar so that was it for the the the mixing of laryngals and and dealers now the non-contact of apparently homoponous phonetics shusler admits that the failure of apparently homophonic phonetics to be found in spelling variation of the same word that the term for this kind of spelling relation in chinese is tongja is a needed explanation and i'll say he's going further there than some of the critics of baxian cigar the people in meijing should say oh one day we'll find a document so shusler said yeah this is the problem yeah but he offers an alternative account uh to that the dealer versus uvillar distinction proposed by baxian cigar namely he says that gong word was reserved for simple k words whereas gong uh uncle uh had complex initials with clusters such as the word uh we translated uncle itself uh which would even reconstruct such as kong but he thinks it's not not guh versus guh is guh versus guh okay in effect shusler here proposes to modify the sheshan hypothesis so here's the new sheshan hypothesis to appear in the same sheshan series it is not enough for initial stops to be homo-organic but they also must be homogeneous with respect to the complexity of the initial if he would have done well to argue that a broad principle of separating simplex and complex onsets into this thing series is discernible in other cases the parameters for his revised hypothesis are also quite unclear but he doesn't say how like okay he's rejecting this sort of standard sheshan hypothesis he doesn't say like this he's alternative is yeah countless series ride rough shot over the presence or absence of a medial r clearly medial r is ignored in in sheshan series so why was medial l treated totally different than medial r so baxian cigar objects to uh shusler's proposal uh as uh follows the uncle phonetic itself is regarded by that's to say the picture yeah is regarded by paleographers as the original graph for this one which starts with a vowel stop under any reconstruction system the middle chinese glottal stop initial cannot reflect an old chinese vealer nor is there any known process for turning vealers into glottal stops this implies that the middle chinese uh initial in uh follow the prince is secondary they're saying this this the character that we use to write things was originally a drawing of jar and the word for that kind of jar starts with a vowel stop so you so your k l trick is not gonna work that's the the point yeah the drawing together five words written with characters from the sheshan series built on the uncle character cigar right as followed then you might want to sort of look back this is this that second chunk of table one now so shusler's forms close to middle chinese do not explain why the gong phonetic was selected nor do they give substance to the lexical root common to the first three items father father and law old man or to the last two earthen jar contained now i will say that the baxian cigar relied in this way a lot on what i call etymological speculation it's not a term they're fond of yeah uh but but uh i would say sometimes it's more logical sometimes it's less plausible but but if there are three words in a series that mean father father and law old man that like are somehow probably pronounced very similar i don't think it's crazy the thing that they're morphologically related right um but i will admit that that when etymological speculation becomes crazy is quite a hard line to draw in the sand yeah uh okay so uh and then they also mentioned that one of these words re-spell that basically after a sound change and they say like well what if if if they were sort of totally equanimous about kind of how closely two things had to be pronounced to use this particular phonetic why did they re-spell it after a sound change yeah okay so um all in all trichler's alternative explanations for the absence of tongcha contact between apparently a model for this phonetics is less effective than the uvular hypothesis that's my conclusion on his attempt to deal with this kind of gong gong problem okay so now onto the foreign transcription so baxian cigar appear not to have rehearsed any of the transpiritual evidence for uvulars in their own publication but they referred to pawn 1997 and polyblank 1962 the main evidence they seem to have in mind is a glottal stuff appearing where further evidence points to a velar so it's similar to what we saw in the shisham series but now in foreign transcription yeah such as the city in koton written umie in the hanshu but written kumia in the hanshu so that's the kind of evidence shisler is again unimpressed and now according shisler according to the uvular hypothesis foreign q becomes all chinese q becomes middle chinese glottal stuff yet if in principle all chinese q can become glottal stuff then it stands to reason that a foreign q could equally well have changed into a glottal stuff earlier when entering old chinese so basically a chinese person uh hear the uvular and understand that the glottal stuff is what he's saying so these transcriptions do not support the uvular hypothesis so i would say one can object to shisler's alternative account by noting that the phonological substitutions acceptable in load-worn phonology are not the same as those seen in sound change these obvious points in particular changing q into a glottal stuff is a perfectly normal sound change but my observation and i don't have like if you know if you have more evidence for this please give it to me my observation is that in lower phonology uvulars tend to be substituted with velars for instance in english where we have words like peran and kibla where it would have been perfectly fine in english just to call the holy book of islam the uran right uh but we don't we don't do it so now i'm sort of done with you know basically the second comment of the bane verwitch is i'm not satisfied by shisler's objections so the second set of conclusions in general should please attempt to offer alternative explanations for the relevant phenomena or unvinvism although to the extent that they suggest a program of research they are doubtless useful yeah so like we should look for instance more systematically at these at the relevant uh foreign transcriptions i would say uh so in one move backstreet cigar explain the phenomena of a laryngeal velar contact initiation series the lack of tonga contact between her and the moth and its phonetics and the phenomena of glottal uh soft appearing in the transcription of foreign words where other evidence points to a velar whereas shisler requires a separate explanation for each case so he i mean you haven't read his article but he goes on and on about uvular violation of opens raiser and i didn't speak like okay who's violating opens raiser right uh okay so in order for shisler's account for laryngeal velar contact and phonetic series to be convincing he would need to show that both the muconetics uh once established uh become more and more common in different syllable types in order for his argument about a parent moth and its phonetics not being used uh for each other in spelling variation he would have to flesh out his idea that it's about constant clusters rather than uvulars and uh and uh the i so what i want to do here is say and if he wants to prove that his account of foreign transcriptions is right he would need to blah blah blah but i can't actually think of what he could do but but that may be my problem like maybe he can think of ways to strengthen that argument okay so then as the overall conclusion i will just sort of remind you of my first set of conclusions and then uh link the two together uh when we bear in mind that uvulars must be reconstructed at the level of sign of defendant for reasons independent of chinese evidence and that at least three types of at first blood unrelated chinese internal problems receive an explanation uh with the uvular hypothesis the coincidence that internally motivated uvulars occur exactly when we need them on uh comparative grounds is i think in effect incontrovertible uh reason for believing and uh and the sound and fury that schüttler directs at the uvular hypothesis is merely his filial expression of a fondness for all lines on i think that's just okay that's the end of my video