 the course into these distinct lectures but they're topical some are shorter some are longer. We got through the first one about methodology last time and if you missed it I did send a link to a related talk that hopefully covers the same material although in a maybe less well organized way. So I decided I'll start today on the second lecture partly because of the problems last time but I will cover the first piece of it pretty quickly because some of you have seen it before just yesterday but you know maybe you won't mind. Yeah so our goal is to figure out what I'm calling using this term of Nikhira Tatsuos the Sonus Gramai of Chinese which is what sound system is implied by the writing system itself and I gave this example yesterday of you know the spelling of the word English enough that would be announced so what is the kind of equivalent of that in Chinese and we use Middle Chinese as the imposer so I'm asking you at this point in the course to just take totally on faith Middle Chinese as if it were philologically recorded in Roman and then I think you know I circulated beforehand this short essay of mine where I basically explain my system of writing Middle Chinese which is you know everyone has to have their own system in Chinese historical phonology but it's designed to be sort of indelogical in its conventions like using n with a dot over it for the Vila nasal and then hopefully kind of pedagogically easy so anyhow I want to use that and now and then the the structure of the Chinese characters themselves are the transposer okay so the character mother so it's composed of two halves one that means woman and one that means horse so one is this semantic component one is the phonetic component and as a native speaker of Chinese of the relevant period you're supposed to just see it and now oh it means something that sounds like horse and means something about a woman right so then we asked ourselves well what is it this actually what's is it this actually been indicated by the character and then we saw that these three characters are all written with the horse phonetic and they all start with M they all have the vowel and but they have different tones so the character is not telling us anything about the tone but maybe it's telling us about the initial and the vowel app yeah but then we look further and we see this series where it looks prima facia like there's two different vowels right there's ah and there's yeah there's also again multiple tones but then also there's different initials so it's not like the character the phonetic is specifying the initial but maybe it's specifying the place of articulation because these are all bilateral stops right so those are the three initials so now looking at this evidence we can form a hypothesis that the phonetic component tells us the place of articulation of the morpheme that we're writing the place of articulation of the initial right and then I won't go through this evidence in detail again but we find that in all poems ah and yeah rhyme so we suspect that actually some somehow they were originally the same vowel that then split in middle Chinese but in all Chinese they were the same vowel so then we can abide by our idea that in all Chinese if two characters were written with the same phonetic uh component then they have the same uh rhyme in all Chinese so in I was using open syllable examples but that would just keep things simple uh and and this is Duan Yu's size observation or hypothesis let's say uh and then in terms of the initials we say that they have the same place of articulation and actually I'll just it's not in the slide I'll just specify that it tends to be stops can be swapped around like we saw at BP and pH or nasals so so it's it's also telling you something about the manner of articulation yeah it's basically saying nasal or not nasal but anyhow and then this is you know back to last time this process of so Duan Yu's side he notices oh things with the same phonetic rhyme sometimes in the in the judging so then you take that as an observation and then you make it a dogma they always rhyme and that's going to be our methodological principle right okay so uh so now we're going to and this is I think we are almost where we left off yesterday we're going to look at the Shesham series that's that's a bunch of characters that are written with the same phonetic and and see when they violate apparently using middle Chinese readings the Shesham hypothesis and then each such violation will inspire us to overcome it somehow with a hypothesis about initials yeah so before getting to that I I want to just avail myself the tool of assigning morphemes in Chinese to two classes and I want this at this point to be very formal like this right we're just assigning morphemes to two different classes and we know these classes exist because uh because in middle Chinese the class V or type B we say has an I in it now you know I you say okay this is circular because I'm writing the I there in order to indicate that they're type B syllables yeah but we have to uh it's you know it's it's a good circularity we have to start somewhere so we're starting with middle Chinese I'm writing middle Chinese a certain way that it is a real distinction middle Chinese we'll get to that but if you just are taking middle Chinese as given as if it were philologically attested in Latin script then you can say okay type B whatever it is is marked in middle Chinese with a medial I yeah uh and here's just an example a type A syllable tongue and a type B syllable piano yeah so uh you know again in the middle Chinese reading okay so this slide is just to now convince you that this isn't a distinction that only exists in middle Chinese because middle Chinese is some dynasty right so it's not all Chinese uh but some not all shesham series some shesham series the bigger ones for obvious reasons because the bigger your shesham series is the more you can use it to make fine phonetic distinctions so the bigger shesham series split into uh uh uh I don't know what to call it a cluster of characters that are indicating morphines of type A and a cluster of characters indicating type B and you see that or one thing that is important about this slide to realize that maybe I should have emphasized more yesterday is there's a tendency in the business even to talk about a shesham series as something flat like this character and this character in the same shesham series as if it's a black and white question you're either in the same stage or shesham series or you're not right but in fact these are nested things right so if you take this little this little guy here you know he is actually quite far from this guy right he uses a character that uses a character that's used in a character that's used in a character as phonetic so um in a sense just intuitively we would expect this character and you know this character to maybe have less in common phonetically than than than a sort of hub does with its spoke yeah and each of these steps is at least providing an opportunity and at this point I'm again just we're doing it very sort of a historically very kind of graph theoretically each step gives you an opportunity at least to make a further phonetic specification right and then we see if we look at the readings here that all of these ones are type A which is to say there's no medial I and all of these ones are type B which is to say there is a medial I right so so we know that this distinction that in middle Chinese is marked by I versus non-I existed already in all Chinese it's it's built into the structure of Chinese characters right so that just means as a as a kind of how can I say if it's something now that we can put things in complementary distribution with because we know it existed in all Chinese you know whatever it was in all Chinese okay so uh yeah so now we're going to improve the conformity of Sheshan of Sheshan series to the Sheshan hypothesis and one of the tools at our fingertips is this classification of morphemes into type A and type B okay so we discussed this example yesterday you look at it and you say well well you know there's all sorts of problems if we believe in the Sheshan hypothesis like we have two different rhymes and we have several different initials dentals retroflex and paletals now you notice actually already you just look at the rhymes like in this first example I gave of the ba and the pie basically we have one rhyme in type A and one rhyme in type B right so the eros none of them have a medial I and the oo's all of them have a medial I right so you can already think oh it was probably one rhyme that split in middle Chinese conditioned by the AB distinction and then you might say the same thing about the initials that in particular you know for for for now the first thing we'll do is say the paletals probably palatalized conditioned by type B syllables yeah so there's what we say we say all Chinese key changes into middle Chinese chuh in type B syllables I will just if you're wondering what's this character it's the name for chuh using middle Chinese so for those of you who who don't want to bother with that just whenever you see middle Chinese Chinese character roman letter hyphen you can ignore the Chinese character but if you're actually you know serious about um doing Chinese historical phonology in somewhat longer term it's probably good to memorize uh which middle Chinese initials have which names in Chinese yeah so what is the question it's um the short answer is it's the head word on the columns of the chai yin re okay uh we'll we'll come back to it but I mean one way you can think of it is just like if you study Chinese historical phonology in china today that's what they would call that initial yeah they don't this is Chinese has its own tradition in this way like you don't say l initial you say lai mu yeah and you only say that about Chinese if you're talking about English or something then you have some other word for l that's translated from a european language okay so uh and then the other proposal is all chinese tr becomes middle chinese uh uh uh retroflex in in any silver i think uh and then uh dr becomes yeah so um and then you know i'm not even sure we'll get to it in this course but these proposals are also motivated by internal reconstruction of middle chinese and and uh like one theme that i'll come back to is this coming back to things right that that the the evidence for or the convincingness of different proposals comes from many different angles but i'm finding i find it you know just in terms of the organization of my own mind and i would think maybe in yours that it's useful to go through these things one at a time say okay now we're not talking about internal reconstruction of middle chinese we're talking about what information is encoded in chinese characters so uh the arguments for these proposals is only going to come from violations of the sheshan hypothesis okay so here was and this is where we did end last time it's another series and i'm trying to organize them to get kind of weirder and weirder right so the first one you're like oh yeah sure dentals and and palados have some kind of contact you know there's some kind of conditioning environment for the palization dental seems quite normal to me right so now you try and use those tools you've just given yourself access to to solve this one and you say okay well then the d can stay and the retroflex probably comes from a media r but then you're like well what am i going to do with this yeah i haven't we haven't given ourselves a tool yet to get rid of the yeah but these all need to have the same place of articulation uh in you know in order for us to win at the game if you like and then the proposal that the discipline has come up with is l right so uh oh yeah so i just went through this actually so never mind about that but uh what so what can become a yaw and a duh well the answer that you know that uh that we think that the field of trance historical phenology has come up with is luck so then let's just go through those proposals uh so l becomes d in type a syllables l becomes yaw in type b syllables and l r becomes duh in both type a and type b syllables and d is excluded because then you would get your jiu so in the first word no it's so this could be a d like like let's yeah this is a good question because it's giving us our sense of a creation of the creation how you use the creation of boxes right so there's nothing this could be a d in all chinese it really could be if we only knew about this character yeah if it were it's only satan series we would say oh it's a d right but how do we get this d and this yaw to make sense right why would you use a character that starts with yaw as a phonetic in a character that starts with duh that is what motivates the hypothesis that uh this isn't a duh in all chinese so let's put it this way some middle chinese does come from duh some middle chinese does come from love yeah and uh what you have uh let's say if i can paraphrase the observation you've made another way if i see a palatal mixed if i see palatals and dentals mixed together in your shijun series then i reconstruct dentals whereas if i see uh i mean stops if i see dental stops and palatal stops mixed together in a shijun series how you reconstruct dentals yeah but if i see uh dental stops and yaw the glide yaw with no paladals then i reconstruct it as a ladle yeah and that's kind of the the the cookbook if you like approach to uh you know all chinese historic monology which is you can you can think of it as a decision tree i have a character that's a d in middle chinese now i need to check does it have shijun series connections to paladals or does it have shijun series series connections to uh to to the glide yaw yeah and actually in my yeah one hypothesis well just to be a little bit expansion in the sense that you have that l r it's kind of yes that you don't you don't like that do you well yeah i agree that's that's all i'll say is i mean but i i i blame myself i say this is my ethnocentrism that i don't like lura lura as uh yeah um there was a time i'll say and this i think is just how in in some ways it's it's actually you know again my overall theme about methodology there was a time around 2000 where we all got very l happy yeah and uh and basically it's like it's like laryngeals in you know european right there's that moment where you're like there's 30 little laryngeals right like whenever you see any problem you just throw a new laryngeal at right so we sort of went through that phase with with l in around you know 2002 and then have sort of backed off and and a lot of the d's have changed back into d's but i think that's this is part of the kind of the the inevitable dialectic of research right and i am also concerned about this lr yeah are we sure that that is close by now no no i mean like are we sure of anything in life i know that every couple of times that's okay oh uh i'll get to that yeah uh in tomorrow yeah sorry i think these are different things so it's kind of like the the top character in the finance of middle chinese and like it's like the what we would pronounce half in like the modern chinese like is this the base or the like yeah you know oh sorry yes um yes i should have well let's say it was intentional it's not a mistake but it was a pedagogical oversight right these two characters are uh paleographically linked and there's no way you could have known that yeah so i i should have put this one up there yeah uh sorry about that yeah last question do x and h rhyme don't we set that all of these have to write right so the final x and h is yeah yeah let's say uh for most purposes um all chinese poets didn't care about this yeah so they do rhyme like um you know they don't have the same rhyme in in a in a precise phonetic sense but they they they do rhyme in the in the churching yeah one way to put it if you like is old chinese had no tones but actually that's that that that's that's a way change phonologists put it um but uh but of course that's sort of just barring from Peter to vape all right because because then you're like well what was that stuff it was still there right it was still something even it wasn't tone yeah but let's say whatever the x and the h represent was not uh poetically uh considered in writing poetry in the old chinese period uh okay are we are we happy with laterals so so far it's just chosen because this this would be the most natural about explore yeah if you i mean if you have a better idea then then be my guest yeah we we want we want something that becomes a d in type a syllables and a yah in type b syllables and then you say well it would help if i knew what the difference was between type a and type b syllables and i would say yes it would help uh but you you get a kind of feel for it right like like and you already have a little bit of feel for it what is a type b syllable well whatever a type b syllable was it changed the densals into paloals and it changed laterals into the glide yeah so it's it's definitely a kind of front happy thing whatever it was right and yeah i know the answer but you gave an example of uh t the retroflex shirt and uh paloalature yeah but similar series exist for the voice the because i'm glad you just concerned and being there in simple elbow wait what do you what i'm sorry i've lost you what do you want to do well you gave the voiceless set for paloals then those retroflexes but voice set oh yeah so yeah this is again it's me trying to be easy right like which is to say like like i didn't give examples uh yeah like why didn't i no i give what is the one oh yeah yeah so i should have said that yeah the only reason i didn't say that was because it wasn't on this slide right because it wasn't in this because this series happens not to happen yeah yeah because i'm trying to build it up very sort of uh empirically in that way but now i see that uh maybe i should have done two series and included one that did have a job but yeah that becomes a job yeah let's see where we okay so now we're ready to move on uh we did this right yeah okay yeah and that yeah so i'm just now like writing it out right sort of writing out our current theory okay so uh this d comes from l in type a sibles and this uh comes from lura in type b sible and so on right and he's just nice too and then i'm not reconstructing the rhymes because we can't do everything at once right okay so now what about oh i give away the answer unfortunately but what about something like this you say uh hung and how can we make them come from the same place of articulation does anyone have any ideas okay you guessed it uh oh no first i'm just gonna give more examples so uh here uh we have uh yeah and then yeah right so you see that there's this there's this connection between huh and uh well both of these are uh velar nasals yeah so so we need to somehow change these hugs into velar nasals right uh and here's how we do it we say they're voiceless velar nasals in old chinese they changed into um into into huh actually into uh i i'm this is my romanization but it's actually an x in ipso yeah okay so um i wonder do we do we feel more or less scandalized by this proposal or by the lr personally i i find the lr more scandalous but you know yeah okay so so good we're on the same page here and then uh yeah so and here's yeah i don't know why i had that on set is slow okay uh and then let's do it with m which is we have uh we have m having a creation connection with huh so we can reconstruct a voiceless bilabial nasal not or rather not yeah okay so you get the idea but i or now i'm sort of like tempted to speed ahead because like now you get it right but i think it's good to sort of put your eyes on each one say like okay uh this one is khan yeah so uh basically we uh yeah how to put it we've it's like we want to get as much juice out of every new proposal right we don't want to propose new segments for every problem we see because then we'll end up with a lot of segments in all chinese so here we don't get i think this is right we don't get her in no shesham series but we do get tough yeah so uh so why not say that the top comes up it's not in type a syllabus yeah and and this is also i think not very scandalous right like it's not becoming tough i think that is a thing that happens around the world uh okay and then what about in type b syllables well uh and this is also where it's really useful to keep track of uh when you see a phenomenon is it happening in type a syllables does it only happen in type a syllables does it happen in type b syllables does it only happen in type b syllables and sure enough you get shes in none series uh and uh but only in type b syllables so then we can propose that snuh becomes tough in type a syllables and shuh in type b syllables and then we also have more evidence of what affects this type a type b distinction uh have yeah and then i'll just point out kind of on saw that here we have nia initial uh which is the regular type b development for an initial nuh which is exactly parallel to tough becoming chow or tough becoming juh right which i also didn't mention earlier yeah well what's the next uh this s with the t what are you asking how do you pronounce that as with oh it's a shot yeah so uh so actually in china they like to write it as a c with a little pale all of it right i think i have a n here right so so so so i would write it like this and in china they like to write like this and this is generally i realize these these actually indicate different things in ipa but in general i would say chinese scholars love this symbol and western scholars prefer this symbol i think our preference for this symbol is not as strong as the chinese reference for this symbol yeah um so so you almost never see shuh and juh in in chinese language publications about any language that's my observation they just they just much prefer if they can to analyze it as this and it usually doesn't matter because there aren't very many languages that distinguish them or i don't want to put it that way instead i'll put it there are many languages that do not distinguish them yeah there are some uh including in china that distinguish them and in that case i think both chinese authors and western authors distinguish them correctly generally speaking um but one reason i like this in the logical transcription so i have to do this test with the shuh is that you know who cares this is we're doing terminology here not connect right there's no way we will ever know uh for sure you know where someone in 1500 bc put their tongue on their on their palate right uh so uh so that's why i like i like these indian um transcriptions because it's one way reminding you this is philology we're doing you're not phonetics right but hang out that's that's my answer yeah uh and actually if you like you know if you want to be precise about this my writing s with a acute accent over it is my way of saying the shuh initial right whatever the shuh initial is which is why i have this transcript uh okay so uh and then we get uh retroflex uh uh so why not take that back to an r you know so basically we're getting all of our retroflexes from medial r at this point so then we have right and then that becomes okay and let me just say the point of this lecture uh is to give you a sense of the information that's in phonetic series right it's not to have to list you know dozens of sound changes that i expect you to memorize by tomorrow right so if you feel a little bit like oh it's too much then just you know go with the flow and uh okay so uh well we we have our voiceless nasals and we have laterals so why not some voiceless laterals right uh and this you know now you see this some of the symmetry of it too right if you look at this series you have yuh and duh you know if you only saw up to here you would say oh a classic lateral series yeah so uh you would reconstruct love oops i should have made this these into a's and b's sorry about that uh the the little pharyngealization symbol you can understand as a as an a that's i was planning to only say this later but that's what back from cigar i think type a syllables where it was for a while uh but i i think that's sort of best just understood as a as a indexing of the distinction in their system but anyhow so uh so then we have these two characters in the in the in the same series and we don't have any solution yet right but you'll remember that that we had not changing into tuh in type a syllables and we had not changing to shah in type uh b syllables and you know i don't know laterals dentals not i'm not as a dental so just sort of the symmetry of our proposal so far would say well why don't we then add uh uh uh shah becomes shah in type b syllable and shah becomes tuh in type a syllables yeah and then we can make this series to fit with the sheshan hypothesis yeah okay so uh yeah so how about a voiceless ra so uh here we have i don't think i know i didn't go into this yet but um but basically uh you already know why we can't reconstruct l uh in middle chinese to l in all chinese it's because we're already using l for something else right because we've said l becomes d in type a syllables and becomes ya in type b syllable so where do the ls come from in middle chinese well they come from r yeah and actually i don't think that this is like in middle chinese l is not contrasted with r in early middle chinese so you could even decide to write these as an r i don't think it really matters very much um but uh anyhow so uh so then we take mechanically we take l back to r and l back to r and then we have this character which is pe in an r series so how do we fix that how do we turn tuh into some kind of r well uh you look no changes into tuh in type a syllables and tuh changes into tuh in type a syllables so how about tuh changes into tuh in type a syllables you know and then we're getting kind of i don't know something some more general proposal like uh acute voiceless resonance in type a syllables become tuh yeah but it becomes chronically federal or what do you feel about then i mean it's closed or in some way and shut yeah do you have any feeling or i don't i don't have a feeling the only thing that's bothered me so far is the l r actually yeah uh okay uh so that was yeah so then uh oh yeah that was in type a syllables yeah uh so then in type a syllables you probably wanted it to become a show right that's that's what that's what you should have wanted by this point but you we don't need it right because there's no shows in in our like shishong series but there are tuhs and in type b syllables only yeah so we can then propose that in type b syllables shuh become tuh so it's kind of like just you know like i don't know like a rubik's cube or something it's like what what all chinese things have i already proposed that will allow me to solve other problems in shishong series right sorry i don't get it is this is there a trend or no no not normal trigger also yeah well no there well there was right this one yeah yeah so so so the only difference here is here it's a dental yeah and here it's a retroplex yeah but uh but this one uh fits the patterns we've seen so far right no becomes tuh shuh becomes tuh shuh becomes tuh in type a syllables whereas no becomes shuh shuh becomes shuh in type b syllables so we kind of expect shuh to become shuh but it but it doesn't or to be more precise we we don't find it convenient to propose that it does because we don't get shuhs showing up in the place where we would want to use that proposal but instead what we do get is tuh yeah uh showing up in type b syllables in shishong series that are basically wrote yeah so that's why we propose this and then if you like you know somehow the roticicity uh was predominant in this case and uh you know uh sort of treated it as if it were a medial r right like i'm being very imprecise here but medial r has been giving us retroflexes yeah so i mean i and i would welcome you to like don't believe any of these proposals i was wondering whether the notation you're using is based only on the two articles that you wrote or like also like the phonetic alphabet maybe i'm not following something because i studied that a while ago but okay yeah so the middle chinese is is what i proposed in my in my article from last year yeah okay uh and then the old chinese i'm proposing like in this room right now right um and it's basically ipa yeah okay okay thank you uh the the the romanization of of chinese characters is we make it to today we make it to tomorrow which is what's in that other article of mine yeah right okay i think any other yeah do you think it makes sense what would be the reason for to use non-reprojects so compared to the books and the record text so you must have something to do with whatever the type ab distinction is and um this this is something i look forward to you writing an essay about uh i mean i don't like i should just say i'm not a phonetician and in a sense neither am i a chinese historical phonologist right so so so i see this is just like you know i'm teaching you the the catechism of the of the church of chinese historical phonology and whether or not you believe it is your own business you know um at least i mean let's say at least in the first instance right like yes we like i think that uh basically that that by the end of the two weeks you should have your own sense of which of these hypotheses are really uh built on rock and which of them are built on sand right and then uh but the other thing to that actually makes it extremely hard to read about all chinese historical phonology is that all of the hypotheses are interlinking in a really intricate way so if you change something then then everything sort of you know it's not that everything collapses but everything is somehow moved yeah so so that makes it like if if there are two people writing about chinese historical phonology and and they 95 percent agree with each other that five percent of disagreement will have all sorts of consequences uh that will make it hard for you as an outsider to read them and think they're talking about the same language right i think that's that's a problem uh and and it's kind of that is the thing that i think is most important to kind of convey right it's like you can't just pick up an article about all chinese historical phonology and read it and trust that the symbols mean what they mean no one talking about other languages it's all about uh these interlocking hypotheses and for each hypothesis either you accept it or you don't right yeah it's nice for you when you are you know you're talking about r because it comes out here right but it's sort of this yeah i mean i would say let's just stick with the main reason for reconstructing r is that it's in an r series right and then you say why am i calling this an r series when the only things in are two l's and a retro text uh yeah and i'll say because i can't reconstruct it as l because i've already used l for solving this d meets y problem yeah yeah but if l is a good like we propose for instance we could reconstruct l here but that would square do less well with the retro actually again yeah that's true yeah yeah would just say i think that the fact that this comes out is retro flex uh at a minimum it's nice it's a nice bonus yeah yeah okay so uh now let's just sum up i think we're going to sum up on voices resonance yeah so this is a little bit of intellectual history so don't come up uh is the first person who proposed the labial the voice of labial natal uh and and then and then uh pulley blank said oh that's a great idea and let's use it for dealers and labial dealers and uh and all this other stuff right uh and then uh baxter is the one who added the voices wrote it so that was there were sort of three major moments in the proposals of these voices resonance uh first comes the m then pulley blank made basically all these other proposals except for the last one we recovered which was the voices wrote it that baxter proposed and and so i actually think that's the sort of exercise we just did although i think i started with the feelings rather than the rather than the labial uh sort of models the history of the discipline in terms of someone comes up with an idea for a specific situation and then someone sees that there's an analogous use that can be put to the analogous idea yeah okay so that's it for voiceless resonance uh oh and then just uh yeah baxter and cigar agree you would follow that because they're they're the uh if you like the the for my purposes the locus classicus of uh those four Chinese today yeah so here's summing up uh voices resonance so uh should all look familiar snar and ma in type a syllables become pa and then in type uh the syllables they also become pa uh and then uh snar and tra become pa in type a syllables uh and then snar and tra become shuh in type b syllables and then tra becomes tra in type b syllables and then this starts to look i don't know natural might be the wrong word but but elegant right yeah uh and then you'll also see that there's a kind of split that happens a lot in Chinese historical phonology between uh what Jacobson would call grave initials and acute initials so labials and dealers are gray and then uh everything else is acute uh it seems to be that this terminology although Jacobson proposed it not for all Chinese but for you know phonology in general has gone way out of fashion uh except in all Chinese historical phonology as far as I know I don't I never do people talk about grave and acute initials but we find it very useful because uh then you have a cover term for you know stuff at the at the basically stuff at the extremes of the mouth is gray and stuffs at the middle of the mouth is acute yeah and and and these kind of patterns happen all the time in the development of Chinese where where labials and dealers work one way and where other stuff works another way okay so that was uh uh it's for uh voiceless resonance and I just want to remind ourselves what we're doing here right we're still we're still just uh looking for problems in sheshan series and then coming up with proposals that will uh change those problems which are middle Chinese problems into solutions for all Chinese right and where we get to believe in the sheshan hypothesis that uh there are a whole more organic initials in all sheshan series so now we're going to come to uvulars and I'll just say that basically the voiceless resonance most of the discipline agrees with I think not in picking although I'm not sure uh but uh now uvulars a lot of people are still on the fence about and then I'll tell you my feeling is the the approach we've been taking methodologically means we should march on ahead and add the uvulars and and I see it only as a kind of conservatism which is if you grew up with voiceless resonance but not with uvulars then you say all these uvulars are crazy uh but if you you know if you grew up like I did with uvulars already in the air then it seems perfectly uh sensible these proposals okay so let's move on to the uvulars some sheshan series uh have connections between velars and glottals or velars and the glottal stop or initial yah yeah so uh so that's bad from the sheshan hypothesis perspective velar should be velar glottal should be glottal right they're different and and yah should be I don't know uh like a lateral series maybe yeah uh so if we have these cases of of non-homorganic sheshan series uh that we will try to fix by proposing uh uvulars okay so here's uh one series uh so we've got uh we've got k and then this dot which is in in uh unicode is called the sinological dot yeah uh uh we use for or I use for but it's not just let's say the fact that it's called the sinological dot tells you that there is a certain tradition behind it in sinology we use this middle dot for the glottal stop uh and then the h is for the ch and then uh then this is my method which which I realized is there are no problems with it uh but it's it's best I've been able to come up so far that's the boystick for more so it's a gamma in IPA this is a yeah uh and you could use the gamma but I then I don't like mixing rick and roman letters personally uh and you know what else are you gonna do you could do a g with a padline or something like that um but I think it's nice to have these as a pair yeah anyhow yeah so you say gamma but that means we could view it like you do right yes these are these are no chains so these are uh so we have velar glottal velar velar yeah although do notice that the velars are fricatives which we haven't seen coming up in velar series so far um now maybe that's just just you know we haven't been together very long but okay and they first are combined here together with a little salt no yes yeah this is the first time you have seen pa and the although I will tell you that the doesn't occur in uh velar series but I think ha doesn't occur in velar series which is to say like um I mean I should put this more precisely if you see a ha you are very likely to see a glottal stop whereas if you see a ha you might see for instance a g uh a gamma yeah which is to say there are two kinds of you know in the same way we had dental series and lateral series we have velar series and we have what I'll call uvular series and what's characteristic of a uvular series first and foremost is the glottal stop that's why it's a violation of the shape of hypothesis right it's the glottal stop that's doing it's creating the problem for us but I will also just point out that it tends to be only in series of that type that the that the ha comes up which is what I'm writing as an h yes you could switch right because if you say to use indigo scripting and you think h is the voicemail right and you could use some zaka then yes uh that is something I have thought of uh but uh yes the thing you're right I think pardon in indigo tradition in well now in indigo they don't use this right they use a dot and they use the dot for the voiceless this thing occurs too not like in invaded or something I don't think I've seen this in Sanskrit yeah that's just another problem is yeah I mean the character is already yeah yeah yeah yeah well so what does it mean in Sanskrit f it means f yeah ah okay well let us h before p okay well the the the reason I use it is because it's used in hittite yeah uh which is not yeah and it's used in hittite I think for for for gamma for x yeah it doesn't matter yeah but yeah that that's why I like it well and and that's exactly why I don't so so I and I'm probably the only person on earth who has this problem wants to write Tibetan Burmese and Chinese like on the same page right so I don't want to use an h of the dot under it now for a visarga and now for a gamma and now for no I want everything to be transparent yeah yeah um and that's you know uh yeah so uh anyhow so uh let's move along uh so we're going to propose that this global stop comes from a queue and now that doesn't get us everything yeah but it's it's a step yeah we're going we're going to make a uvular series and the the the the problem is the global stop so we bring the global stop back to a queue yeah uh in like b syllables at least yeah okay and then uh how about saying that this h goes back to a uh I'm maybe you can come up here and make all these noises for us I think yeah I always feel embarrassed it's quite like I don't know uh yeah no I need to be farther back than that oh I I'm not a phonetician I'll just read it out qh yeah okay so so the x in IPA we will propose comes from qh this all seems reasonable to me uh and then uh the the the gamma how about it comes from the capital G yeah so this is our uh uvular uh idea and you say it will cane it but where does that cake come from yeah sorry this predicts but maybe already said it that global stop was either it's not including good G or no no no you're saying something that makes sense given what I've said so far but I would say uh we also shouldn't have the cake right like where's the cake where's the cake coming from yeah and the answer is worry about it later okay uh so uh so things are starting to look better right oh now now now we're feeling good okay we still have to go with the k but um but but look you know the k uvulars somehow turning into velars is also not a big deal right so so we need a conditioning environment to get the k out of a uvular and we also need to propose which uvular comes from but so you know I mean I think where I feel like this is going nicely right in terms of solving this creation series so yeah why why stop and what to do because you turn it into a great tip right um that's a good question uh I would say probably because the global stop is a global stop yeah yeah I know that the thing is we don't um we like like in backstone cigars all chinese the only fricative is s and I think there's something nice about language with only one fricative that's s right like that's that's how proper language behaves but uh but or or I can make it stab it in another way and say like maybe they were phonetically spurentized under under certain conditions uh in fact they must have been at some point because they turned out to become fricatives uh but it's not a phonological issue because it wouldn't have been contrastive and actually I mean I don't really know languages that um like I don't speak any languages that have uvulars in them not from what I've read like for instance sunda which is an isolate from the fall is analyzed as having phonological uh uvulars but they kind of pop up as all sorts of stuff like ya and various kind of fricatives uh I think that the phonetics of uvulars can be quite glamorous yeah and then the k comes from the uvulars or so yes I mean it has to right like that is you know kind of fricatives because fricatives don't make stops uh yeah good point yeah good point yeah that's a good point that's what I should have said yeah true but that was actually yeah I don't know it's true that that sometimes fricatives can turn into stops uh or under quite yeah okay I'm just yeah there we go yeah I would be very cautious because uh I think you need it for this um I need what well something needs to be thought by there oh yeah yeah yeah that's true yeah well uh but some sort of fricatives yeah okay yeah so for for those of you online we're worrying about whether voiceless residents are fricatives or stops yeah uh in any case yeah sorry uh I mean it's no no no problem right we also have series where uh young is connected to vealers and uh let's try to solve those with uvulars where the main motivation is actually that we that we have uvulars now available yeah uh and we have a conditioning environment which is type b before front vowels right so if you look at uh here what was it it was type that the guck is it type a it's also before back background yeah yeah whereas uh here uh this guck we're using uh type b uh and before front vowel so uh why not use this capital G to solve this link between vealers and uh yeah uh in this series as well yeah uh did I just oh no sorry yeah see the the nice lesson is there's too many of these rules so I can't keep them straight either uh I this is unconditioned it's it's in type b syllables whereas this is the conditioning one condition one which is uh type a syllables before front vowels yeah uh and here yeah here is a nice this is a nice uvular series where you see like you know if you only looked at this one you'd say look why are you tempted to take this back to uvulars at all right like yuh isn't the uvular kuh isn't the uvular but uh in a longer series or a different series here we get the yuh and the thuh and the glalosal and the gap right so this is kind of all the machinery to be proposed so so far for uvulars gets to be used here and in fact one more piece of machinery which is that uh this uh becomes a uh a yuh before um front vowels and uh we've also switched to uh labial uvulars but basically you know the labi what are you what what's the nominal form the label the labialness uh we need because of the labial right in the the medial labial uh and then at this point you could say yeah well why do you want to stick that in the initial why don't you just have that as a medial and that would be a very reasonable point at this point very reasonable thing to suggest this point uh but uh later i will i will get to probably uh that uh we we generally want to put our labials into the into the initial you know okay and now yeah don't let the like don't think of this as a two syllable word there's a one syllable more that i could also put it all over the w yeah yeah yeah if you like the the the b-ness is is indicated with an i and the rounded vowel-ness or the rounding of the syllable is indicated with the w yeah i i like this convention because it is quite legible you know as an angle font you could say ye wen yeah but that's of course not actually what's going on uh okay so uh yeah and then uh this initial which is actually just type b gamma in middle chinese they distinguish for whatever reason the philologic tradition distinguishes sort of usual gamma and type b gamma uh so uh we're going to use bilebial capital sorry voiced labial uvular as our default reconstruction of the origin of uh this gamma in type b syllables and this is i think maybe just since this i don't know it will be the sort of motif what what are the moves that we distrust right the first one i think was this lr this is the second one for me which is it's like the first time we've said and and anything else like that yeah like uh but baxter and sagar uh say like look well i have this this available and i need it sometimes so why don't i just use it everywhere and actually now that i put it that way it sounds like a methodology i should endorse given everything i said yesterday but somehow it you know i it slightly makes me clutch my pearls to say that uh i would really like to have an argument for the uvularness of every series that you make this move in rather than just make it as a default but anyhow they make it as a default um okay yeah and then this hypothesis does account for the distribution of uh of this initial which is only in type b syllables with medial one and then here's uh i think it's a series that will that we will you know use as evidence for the proposal so we've got a wait no we've got a global we've got a global oh that should have been an i i'm sorry there's a typo uh so we've got a global here we've got a global there so we've got our global mixing with dealers so it's a uvular series uh and then we'll take this one back to type these uh labial uvular voice labial uvular and then you see we do get a g yeah uh so there's there's something uh actually we can even probably put a little more flesh on the bones say there's something that there's some conditioning environment that probably takes capital g to uh velar and the takes capital q to velar right but what is that conditioning environment we haven't we haven't seen any reason for it it's not type a versus type b right which is the only conditioning environment that we so far have access to okay so uh summed up on uvulars upon wuyun is the person who proposed uvulars uh he and basically for this purpose for dealing with glottal velar uh links in shesham series but he his proposals are not the same as backs from cigars and in fact backs from cigar uh kind of around 2010 like switched some of the proposals they were making and one thing that you know you you can feel entitled to complain about in their work is that when they change their mind they never they never sort of tell you why they've changed their mind they just tell you why they think what they now think as if they have never thought something else but uh and i think from their perspective that's just like well life is short and it's better to do new cutting edge work than to apologize for mistakes you've made in the past but it does uh i don't know it makes their work slightly hard to use i think that you that you were never quite sure like if you're convinced by what they thought in 2010 and then they're saying something else in 2014 you wonder you know what did i do wrong that i allowed myself to be convinced uh but anyhow uh the proposals that i just talked to you through are their 2014 proposal uh and i will just sort of cheating slightly take a sneak peek at other evidence uh alex satsangha baban uh has an article where he thinks that this which is a these are both sort of titles used by barbarians like uh you know in the haunt dynasty like oh the the shownu chieftain or something so he takes this one back to put it on the sand great ruler uh and then he and then he proposes that this one's related to old tarpid takkan and in both cases we have u-gullers where baxter and cigar have proposed u-gullers there's other problems that you'll see uh on this slide and i don't feel married to this proposal but i mean i do i do think it has the advantage of uh being kind of independent you know researcher from back from cigar uh and maybe suggests that oh yeah there were some u-gullers in some of the places they thought they were u-gullers okay and then here's the summing up of u-guller developments so uh and then i had to break these just the slide was a little bit too big to fit on one so uh in type a symbols you see what we get yeah and then i do both the with the non the the straight u-gullers you like and then the labial labial u-gullers uh and then that's in type a and that's in type b for then we have this little note here about uh we've become disinitial before a or a yeah and that's the end of uh of that presentation oh um so there is this question about why particular characters type a not type b and i'll say it um i like maybe i'm at a stake but it's like how do i know whether something's type a or type b i look it up in middle chinese uh and there are places you can do that like uh um that like a million different home pages which an area for instance will tell you and uh you don't look at type in middle chinese we don't talk about types we talk about grades and type b is grade three and anything else is type a so there's four grades in middle chinese we'll get to this later uh and grades one two and four we call type a and grade three we call type b so that's that's how i know whether a particular character is i look it up on wiktion area and i see what grade is yeah um and that's and that's based on middle chinese books and and right papers any questions could you maybe tell us more about the middle chinese phonology and phonetics then like because instead of like we assume uh i assume just the purpose of this class but like we know how middle chinese sounded more or less like what the phonology of middle chinese also really actually you know like is this everything so yeah middle chinese is basically tested like so um so we'll get to that kind of in the last couple days of the course because i'm treating things chronologically okay so we're starting with the phonetic information in the chinese script because that's the oldest source we have about chinese phonology uh whereas middle chinese we basically have one clutch of sources from the seventh century and then one from the twelfth century so we have a long way to go till we get there uh and you know the the other way you could do it is by sort of having one week on middle chinese one week on the old chinese um yeah maybe i should have done that but i but i i think it's you know i don't know i like old things and and middle chinese historical phonology gets like precisely because it's so well attested i think that once you've been through the kind of a meat grinder of middle chinese you'll come out saying oh yeah his his romanization is actually okay yeah so so then i feel like well let's just use it for a while as a crush uh professor hell yeah i kind of just have the clarification question so on the slide um where it's type b and you know the voice yubu or labial turns into yin which is he otherwise yeah on that side you have a you have a character transcribed into glottostap dos and j so i kind of just yeah that's a type i think it's it's it should be an i the i see i yeah that's because baxter uses j for yuh like right right but i think that's confusing uh so i use i uh and i see and i make it more confusing by having titles that's somehow i'm realizing actually that i should keep track of these things so um maybe send me an email to and then i'll fix that typo and if anyone else noticed any typos let me know and then i'll fix them okay uh last chance for a question oh if i may ask about um so it's not every word with i then it will be necessarily be type b uh it generally speaking it should be uh the one problem is what if the actual nuclear vowel is i and there's you know so if you see di in middle chinese that's not type b so it's um so it's sort of like if i plus a another vowel that yeah if you have i plus another vowel then it's type b and if you only have i you would have to check the initial because because it'll if it's a dental initial it has to be uh type a if it's a parallel initial it will be type b right because we cover that so it's like um if like uh like just like y i w e n then it will it will not necessarily be type b because it's not directly adding wow oh no no that's always anything that starts anything that has i w is type b and these things should all be covered in my my article about my middle chinese transcription although i i realize that you know it'll be full of middle chinese jargon on many of you but you know okay then we'll stop here and i'm glad to see that the technological side of things work better today uh so we'll see everyone in the same place uh at the same time tomorrow yeah okay