 Okay, well Nathan said that he invited three kinds of people and I suppose that I may count as a Force since I fall to the category of random guy that he happens to know I'm not really a synologist by training originally. I'm a japanologist Actually, but I stopped doing that a long time ago. In fact, I stopped doing linguistics for me a long time ago This is actually my comeback. I haven't had a job in the in academia in 11 years so Coming in as a sort of outsider this talk will be somewhat unorthodox in that It'll be in two parts the first part is Is sort of a cautionary and autobiographical story, which is relevant. Don't worry And if you know anything about how Maltese is spelled, that's a big clue right there And this will tie into my remarks on oh Chinese reconstruction in the second half I Prepared this talk Thinking that some of you may not be super familiar with Old Chinese reconstruction and all the decades of wars and arguments about it. So for some of you this may be rather boring But I'd like to provide some background anyway at least from my bike my experience When I first started learning about Chinese phonology in the early 90s My only exposure to it was through the works of Bernard Colburn I borrowed every color in the book I could find in the library and I regarded it as like a religion I totally believed it and In the carbon in carbon and carbon type systems of both Chinese reconstruction There are two kinds of syllables that in the notation I use here, which is not problems one type I Will call after pulley black who came after Colburn a Has no yaw or J and type B has a yaw That is a J in it. I'll be using this color coding throughout this PowerPoint presentation So red means type a blue means type B Now for decades, this is this is the reconstruction that people outside Synology use if you look at Japanology stuff if you look at look at Korean studies stuff You will see that carbon stuff is still religiously quoted today as if Baxan cigar had never happened Which really really irritates me, but that's the way it is And that's that like I said I I originated as a japanologist. So this is this is where I started from I was totally unaware of Baxan 1992 Even as I was reading these books by Colburn. I had just no idea. I mean in Japanese studies. Nobody mentioned it Then I started looking around and I discovered other interpretations amd Pulli black in 1962 Interpreted and in terms of vowel length There have been other interpretation since I'm not going to list them here I mean, I'm not going to I listen here, but I'm not gonna read them. I just want to I just don't you Have to read this. I just want to point out that there's been lots and lots of ideas about what what amd are now 1995 one of my Japanese professors sent me a Journal of the American Oriental Society article by Norman from 1994 and He said oh mark. I think you'd be interested in reading this So I said, okay, and and he thought it was so important that this is pre-internet, right? So he mailed it to me in an envelope to my house and I opened it up and I was absolutely shocked. I Had never seen anything quite like this before according to Norman All Chinese have three kinds of syllables. Yes, you see two here. I know it says three up there bear with me What I've been calling E and B corresponds to Norman's perennialized consonants and his plain consonants Norman has a third type that I'm going to Ignore which is his retroflex class this I only have half an hour to speak so I just can't go into this Barbara says I'm going to concentrate on the French lives versus plain Distinction that I'm supposed so this is 1995 I had just started to emerge out of a whole call gradient religion and This to me just came as just total heresy and unbelievable bizarre and Norman wrote that in his theory his idea was that If a silver was playing that is he has no forage realization it will develop in power medium unless forage realization blocks this Palatine he's argument was based on it was partly based on Arabic This absolutely blew my mind at the time now the crazy things is that I had to take I took a course in which I was introduced to Arab Arabic phonology morphology, etc. So you would have thought I would have been receptive and this is why I say it's a cautionary tale I didn't have to acknowledge but in my mind it was severed from all this East Asian stuff So I didn't want to listen to Norman. I just thought this just cannot possibly be relevant It didn't help also the Norman possibly due to space constraints or something didn't really explain his Arabic argument as much as he could have and so he didn't get any specific example and so to me this just seemed really bizarre and Moreover by this point Since I was emerging out of the whole Calvary in religion I had started to question the whole concept of the yaw being the defining characteristic of the AB distinction and At that point I thought well Norman still believes in this yaw stuff even not at the old Chinese level, but at a later level and Since I was becoming skeptical of this yaw stuff. Thanks to pulley blank. I was starting to switch religions from carbon to pulley blank I was like well, this just can't possibly be right What good could these supposed parallelist Arabic be if almost all this capitalization didn't even exist as pulling black had been arguing so I just totally Put this in a file and kind of forgot about So 1995 So for years, I just played almost fully agnostic about the AB distinction I was certain that it did not involve J. I was not convinced by the other hypotheses like violin that had been proposed Basically, I had no real idea. I just rejected some ideas, but For me it was an algebraic thing Now in 2000 of Wolfgang Bair Showed me a unpublished list of later on Chinese reconstructions by Axel Schuessler. This completely blew my mind Until then I switched religions again from pulley blank to Sarastin and So I was a big believer in Sarastin's 1989 book and its reconstruction In the unpublished version of my PhD dissertation Sarastin has quoted all over the place. If you've seen my 2003 book, I chopped out a lot of that stuff Sarastin's old Chinese reconstruction looking back strikes me as really complicated I actually wrote diagrams for myself to try to make sense out of it what's done to me about Schuessler was that his his account of Chinese historic technology was so simple and It just struck me as currently elegant Here's what I need by elegance Now you may be thinking what why are you talking about? Well Well bear with me One of the key concepts of Schuessler that I just fell in love with instantly. Yep, it's religion changing time again Was what I'll call vowel bending the Schuessler system vowels warp or bend in Schuessler starts off with a six vowel system. That is pretty much John John or Charleston or that's a cigar. So this is the starting point Bending refers to how in doing history of Chinese these simple vowels become More complex in a type syllables and at this point in 2000, I didn't know what a type was. I just said I don't know High vowels bent down And those low vowel lower vowels stayed low. They're already low and you really can't they and it's like you can't push them any lower Congress we can type these syllables and again at this point. I didn't know what type he really was We have the opposite change where lower vowels are called upward And vows that are already high more or less stay high We can argue these phonetic specifics here The point is is that the two different types of syllables are associated with two different kinds of warping patterns The details these patterns are arguable could have varied by dialect and those now What impressed me was that these these that warping or bendings were reminiscent to me of Khmer Now Khmer One could it could try to project Chinese style terminology do not do this But but just for the next two minutes just bear with me In Khmer you could claim that there's two kinds of syllables in Khmer the type a syllables or have voiceless consonants originally and The type B syllables had voice consonants originally in the type a Consonants in Khmer. You're the same kinds of bendings as in typing in Chinese and Similarly the type B now in Khmer again the condition factor is the voicing of the initial constant This is not a factory Chinese so the parallels stopped there, but what struck me as me was the was the patterns of bending So seriously this proposal struck me not only as elegant, but also as Typologically highly plausible. I'm not trying to say that these two phenomena are related in any way the conversion towards my Chinese Or anything. This is just a pure typological Comparison these did not even happen in remotely the same era Which is why I have older versus newer the Khmer phenomenon happened Perhaps Happened with the the last millennium long after this happened in Chinese The point is just that the bow patterns are quite similar and that in both cases you have a You have a two-way distinction that is characterized in the later stage by different patterns of bending now if we take this bending as a fact and The question becomes What makes this happen at this point in 2000 I was agnostic. I had no idea Something was pushing Something was pushing Well, I wasn't I didn't know now What I call the Maltese moment happened in the University of light and light the Lighting University library I was looking through linguistics books there at random one day with no real purpose at all. I Just grabbed some book. They had the word Maltese on it on the side. I Mean this is just completely Thoughtless I opened it up and again put me at random. I Packed to be on the section of phrenology And I pointed out that in Maltese historically where you have earlier Arabic pharyngeal and the earlier Arabic The vows baked In the same way with the Chinese state But less for our purposes for the next couple of minutes. We can call this Maltese type a now in Maltese words that didn't have these The pharyngeal of this you really these that these barrier back consonants You have Chinese like bendings in the other direction. So I spent upwards to year This phenomenon Norman mentioned it in his article, but he didn't say what it was and I had read about a mile a couple of years ago in another article in Arabic But because of the wall in my brain, I just refused to put the two things together The shock of seeing this in Maltese and the Shusei article all within a matter of weeks Just set my brain on fire If Norman had mentioned a specific example in Mali in his article, it might have sold me five years earlier, but No So now they asked what you might look like. I was really excited. I thought wow look at this for intros conditioning so I concluded that That Imala was happening in Chinese and nothing should surprise me because the Malamese bending which is exactly what I've been showing for the last few minutes But at this point I still don't And at this point I started to think now I finally know what a and b is a is chapterized by pharyngealization I Would rather use the vaguer term that some Arabic language use Emphasis because there's a huge literature on emphasis in Arabic And there's this huge debate on whether it involves pharyngealization or eubolization or this or that There's just books on this and I prefer using the vaguer term because when we deal with an extinct language We cannot pretend to write exact IPA transcriptions of what it was so All that matters to me is that these emphatic Consonance is out called them are somehow different from the others the exact details of this was it where they be? the rise or whatever I think Are open to debate? So from this point on I'll mostly use this vaguer term emphasis deliberately Now my next question I'm asking one question where where does type a come from type e being non emphatic? I'm pretty certain of that now. I've been certain of that for 15 years I've been looking into a Into a similar to historical phonology and just done by stuff I see So now the next for the next 15 for the 15 years after this question for me became What was The source of this all this for a realization Is it primary or is it secondary? Has it always been there? So part two now I'm going to get more hardcore Chinese and less biographical. Oh, I was hoping to do that I made a mistake of not drinking anything during the break now We have the key about teas. We have the friend The originalization Symbolized by my key there. So what happens when we lock this door what awaits us next? What's behind the VIP the door? Maltese an example of work on bending itself Now rereading more Norman 1994 with a more open mind that I had years earlier. I noticed this that a two degree a two-way division of words Is a common phenomenon all over the duration continent this made me think rethink the whole question in terms of a real technology now One characteristic of the new back up back since it our reconstruction that strikes the strikes me and Possibly others as very unusual is the huge system of 36 Infatic consonants or fringe alliance or whatever as you may prefer All of these consonants have type B or non-infatic counterparts except for this rare one there I don't know of any language which has more infatics than non-infatics now for some time I was that I tried to Reconceive with fewer infatics than this and I Will confess I fail like I can't really beat this system So I'm going to take this as a given what you see here. I Do know that more or more over the more thought of life This isn't as crazy as I thought it was initially I do know of languages with similarly structured constant inventories. So typologically this is not As bad as it looks When looking into Semitic I found a I found the work of Islam you said who writes a lot about the Arabic of Cairo Now in a description of standard Arabic, you have very very few facts You said description kind of Arabic is stuffed with infatics. They are 23 of them and They have near total symmetry with the non-infatics So superficially this almost looks like the old Chinese inventory However, there is a keyness difference though Most of these emphatics are elephants of non-infatics and not phonies. Most of these emphatics are predictable they are conditioned by the five True emphatics and by and by something else which I will mention shortly The point though is that it is possible to have a language with tons and tons of infatics and it's spoken right now It's just that I I mean this just Probably lose people because of course, there's no way in the Arabic script to write all these emphatics But my assumption is that if you try to pronounce Kyrene Arabic without them, you'll sound really bizarre Norman mentioned Russian and I think Russian put me off at the time 1995 because I because in the Russian the two way distinctions between palatilized and non-palatilized Norman described the Russian non-palatilized concepts as peragial lines. I don't know if I would agree with that but Once again, we have near total symmetry And this time like back since the god of Chinese most most all these pairs really are phonemic But like I said, the Russian distinction about palatilization and old Chinese It is true that some type beat syllables palatilizingly old Chinese, but unlike the calgharinians I don't define type beat as palatilized so there are limits to these palatils of Russian a third example that I studied after I initially read Norman's article was Old Turkic Old Turkic is written in a runic script It too has and It too has a two way opposition, but there is no phonemic opposition between them so the point is that these systems demonstrate that The best of cigar system is not typologically as odd as it may seem at first and I will stop there without actually going into all the original stuff Yeah Okay, yeah now What all the three previous systems have in common is this These large systems Turkic Russian and Tyrone Arabic derive other Diocratically or synchronically from a less complex system In in the case of old Turkic the that is a friend that the two types of constants are entirely predictable in Korean most of them are predictable and It is true that composition is currently coming into Russian, but it wasn't it wasn't like some permanent eternal thing It's secondary So the question becomes then if these large systems come from small systems What about old Chinese? Do we have to assume that the 36 constants are big? I just have forever from sound Tibetan. I don't know Maybe they're maybe old Chinese was like classical Arabic and had a few core In fact and things grew and grew and grew or maybe hadn't known that to originally I give up with hypothesis that In early old Chinese, there were few or no true emphatic and that emphasis was predictable on the basis of something I'll call X In a later stage of Chinese corresponding to the best of cigar reconstruction X is popular Holy God once a conditioning factor starts to vanish Something else may be convenient and in that case it was emphasis So we have this huge constant inventory now these constants in turn influence the following vowels And so that vowel aliphons develop after constant in the final phase of both Chinese Let's say in Han times Emphasis is starting to vanish and the bed vowel aliphons come into their own and become from you And this is then this in turn results in the very complex chain in system of rhymes So what was X? There's another novel trend toward words get toward work structure getting In the in the east and south of station region that's in cigars. We're trying to assess for syllabic Meaning that the word structure is one and a half syllables the half syllable is Maximilious a constant and a swap was the only possible vowel in What's called a pre-syllable? modern Chinese is so-called mouse. I guess we can argue about that, but It's this is pretty much gone and this kind of reduction can also be found other light of use of the area Given that emphasis Was located on the initial cost of the main syllable I thought that maybe X must be near that initial to cause it to become inside What if it so my hypothesis is this what if in fact are traces of lost vocalic distinctions in pre-syllables Which are right before initials. I see in fact step in the progression between dye syllables and model syllables in Chinese language history This is this is now really Here being very dangerous Let's suppose that in very early Chinese you can call it for Chinese pre Chinese whatever They were actual real dye syllables not sesquic syllables and Over time the first syllable start to lose their distinctions and fell into two categories We'll call them Which the type A and B category that type a category is typified by a Low-vowel short up and the type B category is typified by a high-vowel short in in middle Chinese, which is The privilege of the back of cigar reconstruction the short up triggered Emphasis and sure good did not These bottles I have merged into schwa or lost entirely or even the pre-syllable just fell off But the quality of the bow can be determined by looking at the Quality of the consonant Why do I pick low vowels is conditioning emphasis? Some consider low up as a syllabic form of a range of life It's Salish the Fringoluna is a syllabic Alphan of I and in Korean Arabic low conditions for angel alphons This is the most important for our purposes. I think the spreading of emphasis Points with aerial friends. I've been talking about a low high distinction low averses high good There are low high ball systems throughout this region There's a lot of harmony in Eurasia of features spreading I'm gonna skip that and if we looked at Dysyllabic morphine in Chinese and these are a test. This is not my hypothetical Super old Chinese You can see a lot of infatic harmony. I call these type 8E words type AB words based on the type of each syllable My thinking is is that what we're seeing here is the same is the same kind of harmony that would be assessed for syllables But on a bigger scale You find harmony in re-duplication and I've compiled statistics in a 2008 article that I published that I published now it is true that not all dysyllabic words fall fit into these patterns and In these cases, I wonder if they are loanwords or Compounds that have become opaque over time Now I'm going to conclude with some potential issues for What I'll call the extended infatic theory I consider Norman's theory to be the base the original infatic theory minus the extended version the overextended version Well, one problem is this if I can supposedly reconstruct vowel qualities based on infatic emphasis, why is it that we have cases like this where the low vowel that my theory predicts Seems to correspond to a high vowel in a possibly related word This is not good for me. I Couldn't come up with some kind of ad hoc solution to make this work, but I'd rather not this is rather troubling Of course one could just just throw this entirely away and say what all these words is a complete lookalike, but I'm not So sure I Have not by the way 100% endorsing sign of austronesian here I'm just saying that that this could be a loanword. This could be a real word. I have no idea But I don't think this is coincidental And this is not the only case there are others that my theory fails to account for Another problem is if my theory is correct supposedly you might be able to determine vowel qualities of prefixes. I Have failed to do this. I haven't even tried to do this and frankly, I'm a little scared But if I'm correct This should be possible another problem is trying to explain type AB doublets I've come up with ad hoc solutions like this one, but I don't find them entirely convincing To prove this particular ad hoc solution I have to demonstrate that in every case of the body part prefix. I can make it work with a high vowel Pre-syllable and I'll just conclude by saying that I could try to extend my back here even further For the last decade or so I've been doing a lot of work in tundra and I suspect the topic may have also had a type A to B distinction but whether it has a similar origin is something I'm not sure about yet and That's it and I don't think we have time for questions Except maybe at lunch or something