Old Chinese 9: Chinese Historical Phonology | Guillaume Jacques & Nathan Hill | Leiden 2026

So yesterday we spent most of the day uh uh making s prefixes. Yeah. So uh let's just keep it up. Yeah. What do you think? >> Now uh so so far we've had uh uh s clusters as a source of s of of sha and of sha. So now we're going to move on to the voiced uh things. So um so Baxter and Cigar suggested before type B voiced uvular stops the tight pre-initial S assimilated to Z. So here's an example and this is one of Guiam's favorite series. >> Yeah. >> So how do you explain this? >> Okay. Well, I don't explain it completely, but first of all, I think we should mention that there are in these examples, there are two words at least that have a cognitive Tibetan that I think that's significant, right? So, do you want me to write them? >> Yeah, sure. Yeah. >> So, the word for uh young so for for uh obese would be uh so there is a normally you have and look in Tibetan. So for for goats and sheep, but you also have an ancient word >> which means >> you don't need a car. I don't >> Yeah, maybe you don't need a car. So I I read it in >> W. Yeah, that's yeah. So which means a sheep. Yeah. And the car is is maybe and it has for those who do the junk. We just saw the the word the the verb to to be light which is in Japan in Tibetan and actually you have a word for sheep which is >> in the book the white thing. >> No no no no no no the car is just a j thing. >> No but but to be light. >> Yeah it's it's just a senseless they have different tones. >> Oh okay. So it's not it's not it's not it's a metal which means sheep. >> Okay. Really? >> Wow. I didn't know that. Okay. Yeah. So, and then you have this word yang uh which means to be auspicious. Yeah. Auspicious thing. >> Which so which would correspond to >> Yeah. And this is the this is invast right? >> Yeah. >> No. Yeah. Sorry. Another word. But yeah, this is very common for for forpicious is a noun. This is so you have actually two words with a year >> corresponding to uh this uh this initial this series in in in Chinese. So the problem is what happened. So in in so there are several views here some we say these are this is a lateral series and you reconstruct L everywhere that's sus solution and so this means that Tibetan has changed the L into here for reasons unknown yeah first possibility second possibility that bakar saga they say it's it has a u and so the u changed to y in Chinese and basically I But that they say that this represents the uvular. >> Yeah, exactly. They think they they would say, "Oh, look that both Tibetan cognitives have a G prefix. >> That's probably what the uvular became. >> But it well with and you have a u but here is a prefix, right? It's prefix that occurs on well. >> No, it actually matches the Tibetan weirdly well. >> But I mean if you assume if you assume the is archaic and that the Tibetan has just fossilized it, then >> yeah. >> Well, I mean it's an open question, but uh yeah. So we know that it's productive because there's one word that's analyzable among animal words that have this partics. I won't go into detail but and and so uh there's some possibility that you had a primary yacht in or Chinese that the Egyption that's not favored by them >> and that's what Schustler thinks >> that's what that's what I think. Yeah, that's >> and that would mean that the problem with this hypothesis that is a real problem is that we know that this old Chinese yard will become so the yard in middle Chinese right in type B. But what would it become in type A? No idea. >> Well, the other problems though is how are you going to get the K and the zed in this series? Like >> well the K you I mean the zed could be simply S plus. >> Yeah. Okay, >> that's fine, I guess. Yeah, become Z and K could become the I mean yeah it's not more complicated but not propose anyway but yeah no the the problem I see within this hypothesis is >> it means or or what it means that you've neutralize the AB position the the contrast with the initial all change to B uh that's why you don't have a the equival a type equivalent for from for this initial >> and that's a problem also for Beexton cigar that's >> no for for them it's simply a UA series that just happens to >> have to only have B syllables yeah >> so so they're better so so let's just say I mean maybe this is getting a little bit too far into the weeds but why agree with Baxter and cigar because it's elegant >> it makes it it you know because because the the uvular hypothesis in this series is doing the same work as the uvular hypothesis in all series. Yeah, >> there's no special case. So also if you follow AAM's razor then you would believe back and cigar just give me one second. Uh but Guom says and Axel Schustler says uh look if you look across the family the sheep word always has a Y in it and it it you know okay it sometimes has a gene as well but um but [clears throat] it's just not credible that all the languages change their u into yah in this word. Yeah. >> Uh so yeah so that's where we are. And some people feel like look we really want a yah initial in old Chinese maybe also for typological reasons. Um and this is the series that it really that everyone is fighting about. Uh and uh this is where I already you know told you that uh the cigar is going to uh publish about this soon in BSOAS. And actually he I had forgotten but then I noticed the last couple days that he um gave a talk in Vienna on the same topic which is on my YouTube channel. So if you want to see cigar's >> you know donunciation of uh gujac it's not quite true. Yeah you'll be disappointed. It's [laughter] they get along extremely well. Um anyhow it's on my YouTube channel. So let's uh continue. >> Oh. Oh yeah. Sorry. Yeah. Yeah. >> Means sheep. >> Well, you have the other means auspicious. >> Oh, the other one means auspicious. Yeah. >> Like uh today is an auspicious day to get married or something. Yeah. Um uh and then I will say just because one of my favorite English words is psychop. So I will I will mention that uh this is the sheep that is the psychop sheep. So you say what's a psychop sheep? The psychop sheep is the sheep that you you sacrifice in order to help carry the soul to the correct place in the afterlife. Or maybe not carry the horse will probably carry the soul. The sheep I don't know is there >> for moral support. for moral support. Yeah. Okay. But so in Tibetanerary rituals, uh the very earliest funeral rituals, you you kill a number of animals uh to accompany the deceased in the afterlife and the and the the the English word of course coming from Greek for that is psychopus. So carrying the soul. Yeah. So we have a psychop horse and a psychop sheep. And uh in old Tibetan they use special words for these which are young usually. They also use luke the normal word. And then also uh the word for horse is this ram which is an older word for horse whereas the normal word for horse in Tibetan is yeah anyhow that's enough about Tibetanerary rituals but that actually is what join's project in Dublin was about is old Tibetan funeral rituals. So anyhow uh yeah so Baxter and Sugar also reconstruct SD and SMT as sources of Z but they do so not using Shang evidence. So I don't talk about it here. Yeah. Okay. Now we also have this. No. >> Yeah. Uh I don't know how to say it because Sanskrit doesn't have it. Yeah. >> Um Oh, and I should have written it as a a zed with a dot under it. Yeah. So sorry about that. Hopefully I will fix all these mistakes but I don't know that I will because I would have to rewatch all the videos which sounds timeconuming. Yeah. >> Um uh but anyhow here is the example that they give. Yeah. And uh and let's say so far this H the the the voiceless V or fricative uv is the only origin they have for it. Right. Mhm. >> So that the moment you get that in a series, it means it's a uvular series according to Baxter Cigar. >> Okay. And then uh >> here you have morphology. >> Well, but these are also the same shong series. >> Yeah. So here is uh uh yeah the origin uh S as origin of of and here is a series uh and in general Baxter cigar believe that SG develops to G but their evidence for the latter proposal does not come from Shinjang series. So why do they think it's it develops to it's because of the the vowel? Yeah, there's a front vowel >> I think. >> Yeah, exactly. Yes. Um so so and that's maybe nice uh yeah in terms of we've seen that you know >> that we we had it for for uvular but uh uh let's say vers and u do special things before front vowels. Yeah. >> So here the problem is really this and in the same series. This is very strange. Normally you don't have that because the tranu that you have in normally is only in natural series but this special case when it's in a cluster yields the same result. Yeah. And uh also one thing that's noticeable here is that the phonetic is on the left. You see that's the reason is because due to this policization the phonetic was not recognized as the phonetic in Han dynasty. So in the Honda dynasty they they they like moved a lot of >> Yeah. They moved the systematically the the phonetic and and the semantic elements except in the words for they didn't recognize >> they didn't recognize it. Yeah. Okay. [clears throat] >> Uh all right. And now how about [laughter] >> I cannot >> oh yeah >> yeah so back cigar uh maintain that a tight pre-initial s before qr gives rise to this initial in medial Chinese and here is an example >> uh in type a syllables >> uh and here is an example in type B syllables. >> Yeah. >> And uh you see both of these have to be u because they have a glottle in them according to back cigar. Yeah. Some people would also like there to be glottle. Actually they also have a couple glottle series. >> Yeah. >> All right. >> So so for these two series I think not everybody would agree that they are related. Maybe I don't know what would be the explanation in other fields but uh here I think the fact that they share the same line still suggest that they belong to the same series. It's it's and it would be very difficult to explain this cluster this >> well I mean let's say I'm naive so I would say it's just they're obviously related because they have the same phonetic right like so what's the problem and I I would say actually like this would be an interesting one also for the long le school you know uh uh oh you moved to the other side yeah so confusing [laughter] uh yeah so Uh yeah. Anyhow, but I I I I I maybe Guiam is saying this, but I have the intuition that this is the sort of thing that people other than back cigar will find a little bit too crazy. Yeah. >> Yeah. >> But uh you know, but you do see it's it's following all the rules in terms of >> well, you know, they they have the same they have the same shong series. They have the same rhyme. So, uh okay. Now, let's see. This one is for T. So um so yeah, I'll just read it. So back take middle Chinese su as a reflex of voiceless resonant prefixed with s if there is evidence of connection to words that are otherwise known to have resonant initials whether voiced or voiceless. So here is an famous example. Yeah, this is the famous example. Uh so um the the second uh word is right. Uh and then uh you may not have realized that in uh this word which is chin or something. What is how is it pronounced? Chin >> chin. Yeah. Uh you may not have realized that the phonetic is but if you look at earlier forms of the script it is. And I think there's no disagreement about that. Yeah. >> Uh I don't think there's disagreement about that. Okay. And then um they site uh no shang evidence for the same change in type B cells but they do reconstruct the change in this word uh second which is related to two. Yeah. So uh and >> I think so actually. Yeah. >> Yeah. >> I mean doesn't it have >> no other way around? >> Yeah. Look at the old one. It's just ran with a line through it. >> Yeah. >> Yeah. But that's >> but also the second one, right? Those those two little little thingies on the side. I always took them as just being wrong. >> Yeah. I mean to me they look like ice actually. I mean they do. >> Yeah. Okay. They look like ice but yeah. >> Uh yeah we this is where we need uh someone who knows paleography. Yeah. >> Uh but in any case back cigar make it as a morphological argument and not as a shang argument. >> But you know this is where I would like to sort of uh >> uh make it a shaang argument if possible. And I do think you're right. I mean but actually I don't know. Does anyone know paleography? So yeah here here is the the shai form for for n. So actually they have they have this element also. Uh well it's well it's a bit messy. I think so in in you have and but in any way you have >> wait but you're talking about no. Okay. Yeah. >> Yeah. I want to know about this one. I was so so wary is saying incorrectly interprets the photo complent of plus whatever the thing on the side. >> Oh, so the show thinks it's related. >> Yes. But ancient forms of the character apparently says it's related. Yeah. >> Oh, well then in this case I'm going to believe Shushen because you can never trust Wikipedia. Yeah. >> Yes. But but if we look at the stat of the show, they said just Yeah, it's always >> there are two. >> So are you talking about this character or this one? >> This one? >> Yes. >> Yeah. Well, so this one I I just show you the >> the Jaguen here. >> Yeah. So So you have to believe me. >> Yeah. I >> Okay. >> I mean it's not this is not an idea back from cigar. They get it from the >> the paleographic literature in China. Yeah. >> Uh and I think it's quite >> uh I don't know. I can't draw like I draw this is person, right? Something like that. And then this is uh a thousand. >> Yeah. >> But the how are these characters semantically related? >> They're not semantically related at all. That's why they're phonetically related. >> Okay. >> Yeah. I mean whereas for for for these two uh you know you you could you could say if you were for instance I'm not saying he says that but this type of argument he does the reason why they are uh you know if if they're in the same he might say oh it's because of the semantic relationship >> um but anyhow we were sort of Marion was wondering is this is it not possible to analyze this character as in the same Shan series is uh knee >> apparently traditional Chinese is actually written as me and not a student. >> Yeah. Yeah. Okay. >> So yeah. So then it should be Yeah. >> Well at least synchronically now. >> Yeah. Yeah. Yeah. >> Yeah. Well then I will I will uh change that in a future version of these slides. But any case back to cigar make it as a morphological not as a ching argument. >> Okay. And then here is um uh one for laterals. So here's a this is a really nice one. Yeah, because if you got D in one character and a Y in another character, it's definitely a lateral and then suddenly we have a S in there and and you know and then I don't know why not use an S because you don't like let's say the argument has to also be we don't need the S for anything else in voices lateral series. Yeah. >> Um so that's that one. and they appear not to reconstruct any words with type A. Yeah. Uh and I don't know these are I I I feel like I'm repeating myself here and it will be boring, but I always point out when I see gaps >> even though maybe actually this is the kind of gap you don't need to worry about because how many are there in Chinese to begin with? >> Yeah. >> Um okay. Okay. And then here is uh they have this uh proposal also that su can come from st. Yeah. And here is an example. So uh this this third one is a straightforward. Oh, it should be without aspiration. >> Uh sorry about that. Uh and then the pal comes from a dental. That's totally normal. And then we have this suck in the same series. >> How do we explain it? Well, they say, "Well, look, we're getting we're getting so here from S prefixes." So, let's try and get uh so here from S prefixes. Yeah. So that's a s before th and then we can also get it for uh for sq although they think that it depends on dialect and they give the evidence of uh this word and I treat this as shang evidence because it's two readings of a single word right so so or sorry it's two readings of a single character >> so those have the same phonetic right like uh I think ao is the right Latin phrase to use there, right? >> Um, so uh yeah, so there you go. And then I would say this is the kind of thing where like multiple readings of a single single character is good evidence for dialect borrowing >> like primmaacia, right? >> Whereas uh in general we don't like dialect borrowing as a as as a explanation. Yes, sir. >> Is this also good evidence that their capital A is the same as plain A? Uh it I would say it is evidence of that. I'm not sure it's good evidence of that. Yeah. Um but actually like we can make a prediction already which is the dialect that gives us capital A phenomena is the dialect that changes sq into s. Now I'm not sure that that's going to come up in any other word but um >> because there's so many specific phological experience that it's really difficult to >> Yeah. Exactly. But I mean part of that's just in the nature of the the beast, right? which is like uh old Chinese I mean I don't know I I don't I feel like I'm not going on a limb here did have complex onsets and a lot of simplex onsets so so you will get a lot of sound changes yeah >> um yeah anyhow but why is it that they see this as a uvular I don't know they don't say like uh like you know I don't have another proposal but basically basically this character is kind of alone in its series I think >> so um and there's no glottle readings in the series so I don't yeah >> well uh okay and then uh okay so now how is this one different oh this is the same it's just another example of it uh so uh it appears to be that the reason they think it's a uvular is because of the loan into proto Men. Uh yeah. So I'm just mentioning that. Yeah. Um but I'm not I'm looking at Chang and not at Protoen. Yeah. >> And then I'll also say uh I mean Gio may be in a different position than me here, but uh I find Marta Rattlas's work extremely hard to use. So uh yeah, >> it's easier to use than me. I'm telling you that than >> than just reconstructing it again yourself. No, the other works on on proto >> on proto mongian. Oh, that's possible. I haven't looked at any other work on protoian, but I I there's a there is this tradition in uh Asian historical linguistics that you have two sections to your book. One where you propose all your sound changes and one where you give all your data. And I think this is a very difficult way of presenting arguments. Yeah. Uh and that's why in my 2019 book, I decided, you know, paper is cheap. every single sound change, I give all of the evidence for that sound change right there on that page. Yeah. And and I would say this is in Marta Radliff's work. This is one way I find it very hard to use is she separates her data from her analysis. Yeah. >> Um Okay. I just sort of want to skip these details. Yeah. Okay. Uh and then now let's have tight pre-initial as an origin of uh of of this this retro like yeah aspirate. So they uh say s r you know s before a voiceless r and s before uh uh t hr. And I'm sorry I've left this little that's their symbol for uh type a that little fragilization. and I was trying to get rid of them but I didn't do that good a job. So here are their examples. So this one you know we have a I don't know this shong theories and uh and in type a syllables they think it loses the retroflection. No, the aspiration you're right. >> Yeah. Yeah. Sorry, which would be like the actually parallel to the >> Yeah, exactly. Yeah. Yeah. Yeah. Becoming t. Yeah. So, it's it's not >> so it's nice in terms of being parallel. Yeah. >> Uh okay. And then uh here's the other example uh which comes from two readings. Yeah. Uh and then I am just a little suspicious here because they I say if s is a morphological prefix one would expect it to contribute something to the meaning and conccominantly one would expect that the reading straw uh would show a meaning derived from the meaning. Right? So, so here you just have uh like basically two readings of the same character that mean the same thing >> and they think one's derived by the other one with an s prefix and that just like >> that doesn't make any sense to me. Yeah. Um I would actually say better to understand this as having the same protoform and it being dialect variation since it's two readings of the same character, right? But this is the sort of thing where like um I don't know some philology would be useful, right? Do these two readings of of the same character actually have the same meaning? Yeah. >> Um >> this is a surprisingly difficult question to answer because basically our only source of for these questions >> is the jinger. Yeah. And then the question is does this character come up in the jinger? Yeah. >> So >> you can also use dialects. >> Yes. >> To solve this if this word occurs in any dialect. That's more difficult. >> Yeah. Um uh but any I do want to just uh say I do think that are sometimes you know like they have like two words for frog that with different prefixes and I'm just like okay how does mrog differ from the frog right you can't just say h it's much of a muchness. Yeah. Um, okay. Prefix S giving us Z. >> Well, uh, and you see how I'm I'm I'm starting from the relatively obvious, relatively common ones, and I'm going to the more and more niche. Yeah. >> So, uh, well, you this is a really weird, uh, Shashan connection, right? You have one character read Pong and one character read Zang. Yeah. So, this is how they proposed to derive it. Yeah. And I don't know >> you may not find it very elegant but then I would say how do you connect pong and zong? Yeah. Like >> y >> uh and then same >> I think that the reason would be this just a chance resemblance >> trans resemblance >> for for the people who don't think that these are related >> what would you say what would be if one is cool the first two how will they be accountable for >> so so the two first characters would consider to be >> so why is the first character the phonetic and the second character I mean, look, you're not Wang Lee, so we don't need to put you on the spot, but >> if you >> Yeah. >> Uh, and then uh this second one is is interesting. So, the symbol like the character now, how can I say the character I should just write? I don't know. Could you all follow, right? The character that I've written in Oracle Bones uh is uh this, you know, this in Taiishu. Yeah. So, it means me. Yeah. >> And maybe it means me because I point at my nose when I say me. I say that's me. Yeah. Um uh and in the early script, you didn't distinguish the the the character for nose and the character for me. Uh but uh later they um they write uh nose uh as you see it you know with the >> phonetic element >> with the phonetic element. Yeah. So uh actually now I realize if you buy my thing of pointing at your nose maybe they're phonetically unrelated. Yeah. And it's a semantic. But uh if they're phonetically related >> and in a sense they are two readings of the same character >> then we need to connect you know B with uh Z sorry with Z and um this is how they propose to do it and uh one thing you can say about their proposal is that you do get both of these cases with the same proposal right so it has a certain power to it it's not it's not purely uh putting you know epicycles on on your whatever you put epicycles on >> your orbits. >> Your orbits. Yeah. >> Yeah. So I wish to so the the some people are puzzled with the metaththesis here with B be >> yeah I don't know is anyone >> so maybe I can give one example something similar in a related language. So in Japuk we saw this morning the verb to learn which is >> how do you say to learn in >> so is the the prefix that it gets right now >> I can't >> so in Tibetan you have right >> it really means to study and it's from something like So in Tibetan it's the past tense. >> Yeah. So you have a sound change that has occurred in in Japan was long becomes. >> Yeah. Just to show that this is easy to find in a family languages that have the same kind of metastasis. >> It's funny. I I think no metsis bothers me. I just think like you know [laughter] >> metaththesis is a thing that happens you know but >> has it >> metaththesis >> um so somebody says in the chat >> somebody says in the chat is it impossible that Shia could have been heard or confused for apparently it can be confused in Germanic languages and Japanese repeat the question about which >> it was about like a couple slides earlier Shia could have been heard or confused as a as uh here. >> Yeah, I think that's perfectly possible. I'm not sure what what proposal the question is. >> Oh, it's about the word to to write in. >> Oh, it's about the word to write in. Yeah. >> Okay. Yeah, >> I think it's fine. I don't think it's [clears throat] >> okay. Now, I have a concern. >> The S pre-initial does uh these various things. Yeah. Like no, how can I say ts t and dz are parallel, right? Like sz. Yeah. So all things being equal, if they derive from s prefixes, I would like them to be somehow parallel. Yeah. But the first one is from and the second one is from and yeah. And the third one is fromb. So they're just totally different. >> So I feel this is let's say a moment where what did uh I always think about what did France say about the Iraq war? You know this would have been a good time for the Romanians to have kept quiet something like that. Yeah. >> Uh [laughter] I think like this this is a lost opportunity for elegance, right? If they had done it differently, it would have been more elegant. But that's not to say that I have any concrete proposal. But I just want to point out that that lack of parallelism in the origins. Uh so there you go. Now what's the distribution of pre-initial S is also not very elegant. So we don't have SP, we don't have PH, we don't have SKH or S DZ in type A syllables. Although they do reconstruct SB and SK. So this is also like a bunch of gaps that uh and now I think that what they would say because I have asked some questions like this to back scar is you know in a lot of cases these things probably did exist but then were just lost or >> they didn't find it yet >> or they haven't found them yet. Yeah. >> Yeah. Uh >> so so actually this reconstruction predict that we'll we should eventually find example of these and if we don't then we have this pesky gaps. >> Yeah. Yeah. >> That need to be accounted for. Yeah. >> Yeah. So maybe you should all just walk around with uh you know these clusters in your mind looking for >> other possibilities >> problems you could solve with them. Yeah. Uh and they they also don't mention uh reconstruct SMR, SNR, SLR, STSR or STSHR although they do reconstruct uh S voiceless LR and and yeah okay uh so it is clear that there needs to be a lot more work done on old old Chinese phonactics. So if anyone wants to write a dissertation on old Chinese punctics, you know, uh >> yeah, [laughter] >> the reconstruction will change in 50 years. I mean, >> reconstruction's already changed. They're already on a new version that that is accessible only on their database. Although actually they have they're re very recently in the process of uh sharing that database more widely. So uh maybe you know maybe next year I can replace all my these reconstructions with new reconstructions. That sounds like a pain. [laughter] >> Yeah. >> Uh okay. So >> you have painfully learned 2.0. Now you have 3.0. >> Yeah. >> Uh but actually it's their their new reconstruction is I would say mostly fiddling with uh specific words. Yeah. And then a few global changes that uh that are mostly notational. Yeah. >> Um okay. Will you please switch my presentation? >> Yes. >> Uh and then if anyone has any questions, this is a good opportunity to >> to to question. There is similar example in in like the word for bird was like that in was made with the word for word. So that the changed to the after >> is that actually what happened? >> Yes, that is a sound change in Latin. >> They changed what B into B. >> Yes, they changed and after >> after Oh, that's cool. >> [clears throat] >> I think it's related with the founding of Chinese. >> Yes. >> Oh, for red. Yeah. Ruber for Rufus. Yeah. Yeah. Yeah. >> Yeah. Okay. I mean uh Latin I just have to feel compelled to say you know has this wonderful sound change SR to BR right uh and once once you see this then you think anything can happen >> this is this is pro >> yeah what a language >> um to okay So um these were our pre-initials and last time we did the the overall plausibility for pre-initials and reconstructing tight pre-initials using shung series. This time uh we will we will do these other ones. >> Yeah. So we did s now we're doing the other ones. >> Yeah. Just so among the things we saw >> uh because the this yesterday in Jeruk class we talked about the S prefix for positive >> and maybe you have talked about it in Tibetan I don't know >> no >> no because um so actually you saw one of these rare example of sositive in the slides would be an example it's not completely obvious from the middle or from modering reading because they are muffin and even middle reading because one is the G and G. >> Yeah. But so if you follow the reconstruction where it's G and G then you have this >> do you mind just writing the characters on the board with the reconstruction? I I we I should have he wanted to talk about it when it was on the on the board but uh >> this will be in in Nathan reconstruction something like oh why yeah I guess this way this way. Yeah, I think it's a shun if I'm not uh >> well I'm not going to yeah I >> yeah maybe check it's a function and I should assure so it it may be a bit counterintuitive but here it's a sematic element and this is the phonetic element yeah and this is a t yeah uh following no it will this be a j in your in Yeah. Yeah. Something like and I think it's also so this would be so in in my transcription this would be G and uh a G. Yeah. And then this would come from stuff type and this would come from G with paralization of the G. Yeah. Is it is it function? >> No, it's it's the H. >> It's the H. Okay. For the first one. >> For the first one. Okay. Oh, yes. Yes. Yes. Okay. For the second one, I have a look. >> Geese. >> Okay. So this maybe so it's it's so in fact it's a circumfix. >> Yeah. And for for the other one it's both X and H. >> Oh two readings. So yeah actually you have a circumfix like that and actually you have the circumfix also for other examples. I'm going just giving a few other examples because there's a question on on on morphology and I think it echoes what we've done in in in Tibetan. So it's important. So you have Yeah. So this word has two two readings in middle Chinese. One is the ancestor of modern. So it would be like that in sense notation and another is so I think would you write it like this also maybe? >> Yeah I think so. >> Yeah. So so this is to >> so it's like it's not written like that anymore in in Chinese. It's a yeah okay so it would be so it's written differently that's not important so the two readings exist in a show anyway and so this one the here is a type of lateral and but maybe we'll say that today actually so this would be reconstru in type B and here this is actually s maybe similar or similar. So maybe there is a mark here C and this with an X. So this is to fit and here again you have this circumfix S prefix Suffix just like here and that would be the S element here would be the cognate of the S prefix we or the S etc prefix we've seen >> so just to show these examples very few in Chinese I have to insist on that unlike in most of the language of the family they are very plentiful but so Yeah. >> Yeah. But don't you have um wasn't this like m and s also an example of the causer prefix? >> No, that no you have this there there's one or two others but really it's not so conspicuous. >> Yeah. Okay. >> Yeah. Okay. So let's uh move on to the pre the p the p. Now actually I will say having seen this bong and zong in retrospect I would have maybe said maybe the B one comes from but yeah >> yeah that's why not >> yeah uh okay so um >> yeah here's a here's a shong series >> uh so uh you know we have a P and K contact so maybe we can just uh stick a P in this is a very sort of corrant a solution. >> Yeah. >> Body. Yeah, maybe it's right. So, we uh >> Okay. And then uh here's uh where they they put a P before the uvular. So, it's a it's a it's a it's a H, right? So, it has to come from uvular. Uh and then we have in the same series a uh pang. So, uh maybe it was a P prefix before the uvular. Yeah. Uh and then here we have this um I know. Yeah. Uh and uh they reconstruct that. I mean they have several possible sources of that in their system. The one they use is this voiceless R. Uh and then they reconstructed PR. Yeah. With the voiceless R. Yeah. Uh okay. That was it for P's. Not very many P's. Yeah. Uh now on to K prefixes. Okay. So um here we have a nice lateral series, right? We have the the D and the Y in a lateral series and then we have the TH which comes from the voiceless lateral. And then at the top you see this character kang. So how do we get a K initial into a lateral series? Well, here's an easy option. You just put a prefix K on it. Yeah. So, uh, that's their explanation here. Uh, and they appear not to give examples of the sound change in type B syllables for, you know, I'm just, again, I'm pointing out all the gaps I noticed. So, they you can notice them faster than I did. Yeah. >> Okay. And then, oh, this is a very famous one. Yeah. Like, what's the relationship between Lao and cow? Yeah. Uh well maybe it's uh having a K prefix although I don't like the fact that the that the R has switched its voicing uh between the two >> uh which they don't really comment on. >> Okay. And they appear not to give examples of the sound change in type B syllables. So you know so maybe it's like I don't know maybe K's prefixes were dropped in type B syllables. Yeah. Um who knows? Okay. And then actually this is one of my least favorite things they do. Uh K prefix before a gall stop. I was talking about it the other day actually. Right. Yeah. So uh they so they normally reconstruct a series that shows sh contact between beers and glottles as deriving from inherited u as you well know. Yeah. But they also permit a pre-initial K before glap as an alternative explanation in some Shaun series. and this is their example and uh I don't like it at all. Personally, I would much rather reconstruct it with a uvular and then the and then and then and then we and then we've reduced this to the situation we've seen many times before which is why are verers popping up in u series. Yeah. >> Uh I think this is a place where they're really giving themselves too much machinery. I don't know why they do it actually. uh >> because they think that these two words are semantically related and the first one is the nominalization and the K prefix would be a K nomalization >> nominalization. Okay. >> Yeah. That we've seen in Yeah. >> But you could still have it come from a Q, right? You could have a K prefix before a Q rather than K prefix before. >> Wouldn't it yield if it was a learn Chinese? Wouldn't it yield a >> Yeah, it would. Yeah. So, >> so, so, but but in any case, I think the C prefix as a for nominalization does exist elsewhere in the family. Um, I personally prefer this over the other option of the u option because I find it phonetically so so to explain when in the same series you have cur and and gloal stop. The other option is to say that you have a u that yields total stop and that you the the u preceded by a prefix changes to to >> okay I haven't talked about this I haven't talked about this but if he's going to talk about we have to talk about it >> so uh let's just remind ourselves what a normal uvular series looks like yeah so you have uh you have some hes you have some you know and I'll put them in IPA too right this and this Uh and uh you have uh and you have a glottle. Yeah. And then you have some some dealers. Yeah. So let's just remind ourselves where things can come from. So this can actually just come from a gene as well, right? Uh or it can come from a >> a go. Yeah. >> This can come from only a Q in their system. Uh, >> uh, yeah. Yeah. >> Uh, this can come from either a Q or a gal stop. >> Yes. >> Yeah. >> And then this can come from, this is what we're actually talking about here. So it can come from a K or they reconstruct a unknown prefix plus a U viewer. Yeah. >> Or a K followed by stop. >> Yeah. Yeah. Or a K followed by stop. >> Yeah. Uh so uh in this series, what stories are we talking about? >> Software. You're just saying >> that's what we're just talking about. I'm I'm not very happy with it. Yeah. >> So what GM is suggesting is >> Okay. So what they think happened here is and you can sort of uh I mean I don't know it doesn't seem entirely implausible to me. Suppose we had something like Yeah. A >> they're suggesting that this pre-initial was the conditioning environment for this changing into and then the T drops and you get Yeah, that's what they think happened. >> Uhhuh. But uh they because precisely because it dropped there's no way of knowing whether it was a T, a P or C. Yeah, >> that's so they write this with a capital C. Yeah. So I haven't got to that. I'll get to it soon I think. >> Um but what Giam is suggesting which which I do think is elegant as an explanation in this series. So uh so Oh yeah. Okay. So it's it's this person. Yeah. It's this person like this and then it's the person next to some little thing, right? Yeah. Okay. Um, so this one uh they do as o and this one they just do as o. So he's saying this is a k prefix. Yeah. >> Or and back cigar are saying this is a k nomalizing prefix >> and actually can you think rapidly of a Tibetan example because they >> for for for the K prefix or for this >> for the K prefix? Well, for the K prefix in Tibetan, you could look at um >> I have >> Yeah. Yeah. Yeah. But this one is a bit maybe more complicated. For examp of of the powder >> like actually you say like to be careful. So chip. Yeah. >> Um Yeah. and gip >> and then gip >> the fine finely ground flour you know >> is a certain kind of flower I don't know this word >> this this is in ultip very not very >> common >> uh anyhow uh and uh in Java code k prefixes >> completely doesn't yeah >> it's it's like all verbs take it whenever necessary >> uh so uh so Gon thinks this idea that this the idea that this reading uh gets normalized as this is plausible. >> So it would be like the qu the the >> fearful >> the fearful one the the frightening one. >> Yeah. >> I'm not I'm not completely enthusiastic with this but at least it makes sense morphologically. Yeah. >> So but what is stopping us from reconstructing C. Q and then the C maybe. >> Exactly. That's the problem. You could be a better option. That's exactly what Okay. Yes. Yeah. So what >> nothing is fluffing up from. >> So yeah. >> Yeah. Well, that's what I was saying is like they could have just reconstructed this as uh CQ >> and this one as you know >> Q UJ. That's >> and then just have a strong opinion about what that C is and say it's probably K. >> Oh yeah, you could do that. Yeah, you just say in this case we know it's a K. >> Yeah. What do you think of that? Well, I mean I don't like pronouncing but plenty of languages are okay with I just I just don't know if uh you have any other onsets where you have a viewer and you like this. >> Uh no only one of the language. >> Yeah. Yeah. Yeah. But then you might need to examinate the uh the cognates to see what the um >> we don't know cognates of this. So I'm from my perspective the problem is that don't expect a prefix a cluster to change to >> it seems to me very weird >> uh unless someone finds an example and shows it to me. >> Okay, so let's just do three uh three proposals. >> Yeah, this is fun. [laughter] >> I hope the rest of you are having fun. So we have Baxter and cigars, but they have for in this case they have this versus this. >> Yeah. >> And then uh No, sorry. No. Yeah, that's that's right. Yeah. Uh and then uh Marine is saying uh look uh just do this and uh we'll do it. Uh and I hadn't thought of this as a solution but it is definitely in the spirit of my thinking which is like basically you don't want to introduce more machinery than you need. Yeah. So they already they already have this machinery. So, let's just use this. Yeah. >> And then Guom is saying uh that uh actually it don't you don't really care actually as long as you get a K prefix. But uh what he disagrees with is uh or now now yeah let's just now go over here and say okay in a series and I'll just make it a uh where we have like u you know ca and a and uh >> and ha >> and yeah and y Uh what Baxter and Cigar would do is is they would do uh CQ uh >> Q QH capital G and capital G in type B syllables. >> Uh and what GM would do is the same in all these places. Yeah. But up here he would just say why not uh KQ. Yeah. >> U I'm not too sure even KQ would that well yeah I don't know. No actually that's a problem I have how how to account for this. So Puin said that villas are just >> would be ju just phonetically close enough to use vas that you can mix them >> that that Nathan thinks that is abandoning the scientific >> Sure. I'm sure >> I mean you can also say that for any series right you can say and are close enough that they're >> Yeah, exactly. At that point, >> you could say that and are closed. >> It's like a Pandora's box. >> Yeah, it's a Pandora's box. That's why I don't like >> um but uh here we have a question that actually any of you could help with which is around the world under what conditioning environments do u change into viewers? in the dictionary >> when they don't disappear totally in a language because in some cases you can have a language that simply loses the emerge with villas >> but when they don't completely >> okay when let's do it this way when when some of the uers become gls what >> is there a conditioning environment where the the same u also become Yeah, >> but you don't need that here, right? If you have a initial C, you could posit that it's just K glottle stuff which like like the V drops and you have K glottle stuff after the be. Yeah, but then you have like >> Yeah, but just for like a single generation then becomes C. >> I think it's funny. >> Yeah, maybe. >> Yeah, maybe that works. >> Oh, so so Josiah is proposing that we have something like and there's no one would have any problem with that. >> And then it becomes >> and then it becomes >> Yes. And this might have be for five minutes. five minutes but I mean >> that's Jay Jasnoff always says this for five minutes >> completely really unstable but like this is the kind of change that would be quite natural >> and then this becomes ca >> yeah or or you could say that that the vowel drops out and then you have a a ver uvular cluster and then there's like a harmony going on where you have the uvular becoming a but I think this is actually the best explanation right here >> however the this explanation would predict that if the the prefix is not cur if you have a t prefix >> this is something I've been wanting to say have >> yeah exactly then you have a third >> but you would expect these series to have as many T's as K's >> yes >> and they they >> don't although actually no one's kind of looked at the problem with this in mind either right which is maybe this is how we can get >> our you know because we have DQ you know and we have >> I don't know like it it becomes >> that's my gut instinct is is that then Tuka should become just >> yes star yeah That's >> uh and my gut instinct is in the Medine reconstruction we end up with too many K prefixes in uvular series but it's actually an empirical question right like maybe >> it's okay yeah >> just of the short side note of of and connections in Quranic recitation you have one reader who reads as off when it comes to you get a list on the general cap >> which I am happy with. It's the other way around. It would go the other way around but >> you know but you get the time and get there that way. >> So >> all right I feel worried that some of the students are are thinking about you know running through the the fields the summertime or something. >> Yeah. Yeah. >> No, [clears throat] but this was really >> good. I thought thought. Yeah. >> Anyway, this is what we all talked about. Yeah. >> Yeah. >> Yeah. Here's the inclusion of readings with a with a or a y mean we have to reconstruct the u right. So there are there are very the point I'm trying to make here is they allow themselves K prefix u u uh glto stop >> but uh most of the time they can't use it because there's actual evidence of a ubvulum right because if you have a h or a or a >> then you have to reconstruct as a ub yeah >> and then I will say that I don't think it's a coincidence that it's only in a very very short series that You know like basically if a series so so my instinct is >> probabilistically if a series is long enough then you have to do a uvular which is why they only reconstruct this K before a glotop stop >> in this one case. Yeah. Whereas what would be nice to prove them right is if we had just had a long series which was just K glottle stop gl and then you would say okay that really is different than the the phenomenon that we have with uvular but it doesn't happen >> and then that stop cannot be reconstructed on any other basis except in this case right otherwise it's indistinguish >> uh there are I won't be able to think of them right now but there are series that only have glow spots. >> So, uh they tend not to be very long, but actually I think that's also maybe for for sort of psychological reasons almost, right? >> Uh but but uh uh back cigar are happy to uh reconstruct those as >> okay >> stop is is um is um what is it? Is it the one that means like anger? Is that one example like that? Yeah. Yeah. Yeah. >> Yeah. I think that's one that they reconstruct as a gl and one actually. >> Oh. Oh. E. E. Yeah. >> Yeah. Okay. >> That's a really weird word. We don't have it in other languages. Yeah, >> we do. We do. >> Some languages have it. Yeah. >> Like what? >> Who has it? >> Bur doesn't Burmese have it. >> Oh, yeah. Not as the normal word for one, but some. Yeah. It's a >> yeah and the center word and also that's a lot and that's a lot of words with only words. Okay. >> Okay. Let's move along. This is a t prefix before kr. >> So this is all >> these are some of my favorite ones. Yeah. So here we have uh uh the word for nine which looks like an elbow. I don't know. Maybe it looks like a spark plug. [clears throat] >> [laughter] >> Yeah. Okay. Uh and then you put um meat next to it. And so now we now now I don't know. I think I think this is back on cigars explanation. The original word for nine was the symbol elbow. Yeah. Because the two words kind of sounded the same. Yeah. And then they wanted to say, okay, now we want to write the elbow that's really an elbow. Yeah. So then they added the flesh radical to mean, you know, the elbow that's actually flesh. Yeah. >> Uh the meat elbow as opposed to the number elbow. >> Uh and um uh Yeah. And this is also a case where uh Japok has nice cognates. >> Yeah. So well so >> the detail is a bit complicated but so in you have the so here the elbow >> for yeah so this is the is what in this is a um like a possessive prefix >> indefinite possessor prefix so in some direct of juk you have to >> so not my elbow or your elbow but >> someone's elbow sorry someone's elbow >> someone's elbow so how do you say my elbow that Yeah. So that's in one but in the D we are studying it's a different word as >> which is from mixed with Tibetan but so yeah you definitely have and actually they explicitly quote data from Gangi to reconstruct it this way. So that's a bit the problem if we we can't really use this old Chinese to to confirm what we have in Gongi since Gong already informed the old Chinese. Uh >> I guess >> I would say yine like I think this kind of a circular reasoning is totally fine as long as it's not a pure and complete circle right like >> uh >> because because uh the actual progress of science is always in a kind of spiral movement right >> so um so I think people are too cautious about circular reasoning I think my advice to you is do circular reasoning with abandon and Let your enemies find where it doesn't work. Yeah. [laughter] >> Uh [laughter] okay. Uh so um I mean what what what would be nice about this is but actually what's what's your explanation of the like what how would we what I was going to say what would be nice is it shows that actually uh or it could or it suggests is maybe the right word to use that uh Chinese had this >> indefinite possessor prefix >> and and that it was reanalyzed as part of the root here. >> Yeah. Just as there are examples for instance in in cross a language considered to be languages but has become part of the root because the cross has completely lost this. >> Yeah. >> But it's preserved in a few body parts. So it's perfectly fine. I think >> that this prefix would get lexicalized in a few body parts. >> But let's just uh think this through very carefully. Suppose the word for elbow was the crew. >> Mhm. And now I'm uh who was the guy? Sanja the guy who invented the script. Yeah. >> Sanja. Yeah. So I'm San and I'm trying to think of the how to write nine. And I think well nine is pronounced k >> but I'm like well I can't I don't have a symbol for nine. So then I think well it sounds kind of like yeah. >> Well maybe it was the maybe there was a t prefix. No, no, no. Probably nothing. I don't know. >> Yeah. I mean, I think it's like like in the very earliest stages, you have to be pretty loose with your phonetics cuz you you won't be able to get pictures otherwise. Yeah. >> Uh anyhow, let's move on. Here we have uh uh Yeah, here we you can see it. Yeah. And you can think it through. But anyhow, here's evidence for so we can only get the uh the uh actually so so we could get a ch from a C but only with the vowel it. And this is the vowel ooh. Yeah. >> Yeah. So uh so here we need a >> Yeah. To uh for to uh exit. Yeah. >> All right. And then here is uh the C becomes P. Yeah. And here's a nice long series. So you see, okay, it's K, K, H, G, K, H, R, and then we have this T initial one. And I would again just say like there are a lot of these examples and it would be interesting to know how other schools of old Chinese reconstruction deal with Shan connections like this. Maybe they say, "Oh, whatever. It's not common enough to worry about. Uh but anyhow, now I'm done with the T prefix. Okay. So now just remembering what I did. I did the S prefix. I did the K prefix, the P prefix, the K prefix, the T prefix. And now I will do prefixes before resonance. Yeah. So the best evidence here comes from multiple readings of the same character. So we have this first character which is read pim and limb. So maybe we take it back to prim. Yeah. And then one that's chip and nep. So maybe it goes back to tenepp or maybe it could go back to the Yeah. But uh anyhow uh and then we have to invoke dialect borrowing here like like dialect mixture but again I don't mind we should we should always take dialect mixture as a last resort except when you have multiple readings of the same character because then it really like that's what you expect from dialect mixture, right? Uh and then um and then this last one we have these two rings quet and m. Yeah. So uh which is going to be from kmet. Yeah. So so yeah. So there you go. And then this is a place where where you can really make use of the loose tight thing too. You could say well one one dialect had a tight pre-initial K and the other one had a loose pre-initial K. So in the one it became kept and the other one became you know met. Yeah. >> Uh so I think these are also just in general kind of good evidence for having to reconstruct clusters. Yeah. >> Uh and you see there I don't know you you see why I didn't discuss this under the P prefix or the T prefix. They kind of form a natural class of uh prefixes before uh voiced uh resonance. Yeah. Yeah, I'd like just to mention this to kind of sound Chinese is at >> in languages. >> So in a language >> also in Siri, right? Like Siri is is was historically and they pronounce it as something. >> Okay. >> Yeah. So >> so so you have for instance the the something that comes for for the dog. Yeah. that turns up a in time with in prototype and in you get you know just an example of of that this this type of sound train is not >> yeah sorry is not a odd sound train >> no it's not yeah >> can we check the chat and see if there's anything >> just a few comments uh >> okay >> somebody saying uh apparently Arabic comes from Yes. >> So, so a possibly a uh a followed by might become u again I don't agree with that. I mean I mean I mean I think it's true but it's it is because it is uized right so what is a >> ubized u >> yeah but actually we had this whole discussion uh yesterday at lunch about whether old Chinese might have had adjectives based on kind of >> the fact that adjectives give you u in uh in >> symetic languages >> uh moving along. Okay. So uh so now tight nasal pre-initials M and N on the basis of Shang evidence tight M and N are very hard to find because they just change a voiceless stop into a voice stop according to back. So, so if I have in a series, you know, uh if I have in a series something like pa ba p etc. It could be that any one of these vas according to bags and cigar comes from. Yeah. Uh and precisely for that reason you can't tell using shing evidence. you have to use loans in mon. Yeah. Uh but we can see them uh before uvulars and r. So let's look. Okay. So uh yeah and here I've reconstructed all possible origins in order to not be seen to be taking a rabbit out of a hat. Yeah. So we look at this series and we have uh we have a glottle stop, we have a y, we have a k. So we're clearly in uh uvular land, right? >> Uh and then we have a ver nasal. Yeah. So they say this is a uh comes from a a nasal of some kind before a uvular. Yeah. Which is okay with me. Uh but actually now I've just thought why don't we just reconstruct lab nasal uvular here right maybe there are typological reasons not to but you know what do you think >> I don't >> you'd be okay with that >> I'd be okay with it yeah sure I mean if you have nasal in all the places why not also yeah what do you think >> why not although for this particular series I tend to believe that Francis the the U the give word may may have had a year primary Y because you have a >> then why is in this series with K >> I don't know but well with the K it could be K yeah okay >> with the rest it's not so obvious but because you have yeah to give in in like that so maybe that's could be another give verb that's put >> yeah I am actually fine with saying that all of the Y across all signage didn't come from uvular. I don't It doesn't bother me at all. I don't know why. Yeah. >> Okay. >> Well, I mean then we should question whether or not these are actually uvers, I suppose. >> Why? >> Because u becoming y is not that common of a sound change, right? So >> for a capital G to become Y, it's not that big a deal, right? >> G to Y. >> But when you expect a lot of languages to have >> in type B syllable, >> do have type B. >> Yeah. Well, that's the question. [clears throat] >> So, yeah, >> there's a question. >> Capital Native. >> Capital N means of n unknown place. No, no, not of underdetermined place of articulation. >> Exact. >> So, here it would be uvular actually. Yeah. Yeah. But this is a thing that it actually goes back to who who's who's doing Tibetan with me here. Yeah. Okay. Yes. Yeah. So, uh it goes back to this whole controversy ultimately about the 23rd letter of the Tibetan alphabet because in Tibetan you have a a morpham that I think was always a gamma, but Gom thinks was prenasalization. Yeah. Uh, and I think that, you know, at least it like I'm not making a logical claim, but I think historically the idea that Sino Tibetan had had more that were nasals of unspecified uh place articulation comes from this observation in Tibetan. Yeah. >> Uh okay. And then here is um uh yeah they they think that maybe one way to change an R into a Y is from uh a nasal prefix in front of the R. Yeah. Make it what you like. But in case this data does need to be explained which is this character has a reading with yah and a reading with la. Yeah. >> And they postulate that middle Chinese shows a mixture of two dialect developments. In one dialect, M R leads to L in a word like uh >> Yeah. Yeah. Uh and then in another dialect, Mr. merges with M. >> Yeah. And then actually, so you you do see also that they do like how can I say it's manifestly the case that Chinese has had dialect mixture since the beginning of time. Yeah. So I don't think it's uh too crazy to invoke uh dialectic mixture as often as they do but you would like to find some philological evidence of like oh in this part of the country they do it this way and this part of the country they do it another way and in in China I think you know it's it's although it won't be possible in all cases to do this >> uh there is a lot of evidence in the history of China so uh it might be possible yeah that would make a good sort of MA thesis you know to try to look for everywhere action and cigar, you know, propose dialect variation. You see, is it borne out by the primary sources? >> Okay. And then um uh yeah, yeah, I'm this is me having my normal worries. I was m different than m. And uh and um and and I think that that reconstructing mr uh is enough of an explanation. Maybe why do we need M dot R? Or maybe this is maybe this is just about morphological analysis. It's a little bit unclear to me. >> Yeah, me too. >> Uh okay. And then here is uh we have these this series of this N series and then we have a m coming up in it. This is a fun one actually because I will tell you how Christopher Beckwith explains it. And Christopher Beckwith is someone who makes Gong react like that. But but I but he says in uh Tibetan varieties. Uh I I I want to go out on a limb here and say something like so so the word for person in in Ultib was me. M yi and in omo in some omno dialects at least it's pronounced ni >> n. Yeah. Okay. >> Yeah. >> Uh so the my y has sort of palized if you like just to a normal n. Yeah. So that's how uh Beckwith explains this. He says no it's not an m prefix before. It's a they were all originally m and then uh there was a sound change. I forget how he does a conditioning environment, but I would say this is not a crazy idea. Yeah. And uh and the trouble with Beckwith is he takes lots of very astute observations uh and clever ideas and then he's he sews them together into something that is ridiculous. Yeah. >> Uh so it's very hard to engage with his research, but this is something that I just happened to have read. Yeah. Uh, and then here is another example of an M intruding into an N series. So maybe an MN prefix. And then if any of you say like, no, no, no, MN is weird to say happens in Tibetan all the time. Yeah. >> Yeah. Uh, okay. So, tight pre-initials before u as a source for verers. This is what we were talking about just a few minutes ago. How do we actually explain the verers that pop up in uvular series? So back cigar proposed that tight tightly attached stops you can see what it says that tightly attached stop pre-initials are the origin of verer initial readings appearing in ubular series. >> Okay so here is an example. So uh this is the example that we were talking about yesterday actually this is the this uvular this famous uvular series. So it's gong uh actually just because I have lots of people who who know Chinese well here including Gyongjac. So this is the gong that means public and the gong that means work is the one that's a verer. >> What does the gong in come from? >> Well is a is it doesn't have the same vowel in >> Oh okay. Yeah, it's a it's from a O rhyme. This is No, no, no, it has the same rhyme. I'm just making >> Okay. Well, >> yeah, indeed. >> Uh, so here it's a classic uvular series. You have a a glottle stop initial and a y initial. Uh, and then you have a a K. Yeah. So, how do we explain the K? This is how they explain the K. and and I can't defend it any more than saying this is how they explained it to me. Yeah. >> And then as you saw, Guiam is not very happy with it, but couldn't exactly come up with a better uh answer. Yeah. >> Uh and then uh and then uh Josiah says it's a K prefix and then the question is, do we end up with too many K prefixes that way? Yeah. >> Yeah. >> Okay. And then uh here's another one. So it's a K series in a u series. And it has it's a short series. Yeah. has to be a ubular series because of the >> lot of stop. >> Yeah. >> Or or it could be this is could this could be another place where we want to do K plus uh >> but actually you also have the K which has a K. >> Yeah. >> Yeah. So it's >> okay. And here's another one with the Y and K connection. So we've seen a lot of these before and now I'm just showing you how they reconstruct them. They put a capital C in front of the ubular. >> Yeah. >> Uh so I'll just uh go along, >> you know, give you a chance to Okay, that was it for that. >> Yeah, let's do this one. Okay, >> we'll get started with this one and then uh cuz I wanted to uh basically do what I could with just this is kind of coming back to the first of the start of the week with me. I wanted to do as much as I could with just middle Chinese and just chum series, right? So uh middle Chinese is my imposers and uh >> shan series is my transposers. >> So now I want to just uh get us familiar with the other uh tools in the toolbox. Yeah. >> So can you make the left part disappear? >> Oh sure. >> Uh okay. So cognates and early loans. Uh so these are other ways we can attach uh phonetic substance to old Chinese characters. Right? So I'm going to distinguish loans used by Baxter and Cigar and loans not used by Baxon Cigar. So first is it appropriate to use uh sorry. Yeah. So these are yeah these are the three categories of this presentation. Cognates loans used by Baxter cigar and then loans not used by Baxter cigar. Okay. So now first we do cognates. Yeah. So is it appropriate to use non-synetic cognates to reconstruct old Chinese? Who who thinks it is? Okay, that's okay. Someone count the votes. Josiah, you count the votes. One, two, three, four, five, six, seven, eight, nine, 10. >> Who thinks it isn't? [laughter] >> Okay. [clears throat] Well, here I was planning to convince you that it it might be okay some of the time, but okay, let's ask Josiah, why isn't it okay? because well first of all unless we have a reconstruction we can't identify cognates uh and also uh it assumes the relationship of words which may not actually be related upon further uh examination. Uh we also this is also just not what we do generally right like like we don't um we don't assume like if you're reconstructing a macro family let's say like doing some aphroiatic or something you you shouldn't reconstruct something that is just an intermediate form of all of the the descended forms because even if it works it's extremely ad hoc and doesn't take into account like analogy or >> so so you've you've come up with several reasons that I sort by but haven't quite put your finger on the one that bothers me the most which is just what is the object like what is the thing we're reconstructing right like like if if if we're trying to interpret old English texts we wouldn't use Latin right or like or we would we would of course use Latin >> we wouldn't use lat like >> we wouldn't we wouldn't use Lithuanian let's put it that way >> if was not yet reconstructed we especially would not use Right >> and and Even if we did come up with like there's some there's some character in old English we don't know what it was. Yeah. And then we find out oh it has a beautiful correspondence with this L in Lithuanian or something. >> And at this point you're you're proposing a thing which is not an interpretation of the old English. It's like some some kind of hydra monster thing where it's like I I'm you know it's it's more like protoChinese than old Chinese. So >> it's pre-ro yeah yeah I agree with that completely. So, so let me >> you you seem to you just said that we shouldn't worry about uh uh having making secular reasoning, right? M um so I don't know uh philosophically but so for for for specifically I think there's one area where the use of languages outside of Chinese is necessary that's the interpretation of morphology >> because once you morphology collapses you're not going to be go very far if you just look at internal data just the way that even if you do a Germanic linguistics you pretty much have to look at Sanskrit and Greek to make sense out of your morphology. >> Yeah, the question. So I completely agree with that and I hope everyone does. Uh >> it's very different from using Chinese. >> But can you do it for phenology? >> But no, but I was referring more to phenology. >> Can you what do you think for so for the year? I think yeah because if you just go with the Chinese, you'll construct something that's a monster from the point of view of the whole family. >> Yeah. I mean what I would say actually is I I will I will stick with saying circular reasoning is totally fine. Yeah. And as long as you kind of don't forget what you've done. Like if the only evidence you have that a certain letter in old English was pronounced as an L is a systematic correspondence with Lithuanian. This is extremely weak evidence. Right? So the risk is that you is it is that later you will confront better evidence and you will say well everyone knows it's an L. Yeah because of the force of tradition. So I think >> as long as you uh >> uh >> weigh any evidence. >> Yeah. Weigh the evidence. And actually the way I you know I'm I'm I'm putting mine to one of Jay Jasnov's slogans which has been extremely useful for my career. So I give it to you is never fall in love with your own ideas. Because if you fall in love with your own ideas, you will invest your time and effort in protecting your your bad ideas rather than coming up with good ideas. So, so you should just uh you know like uh like when your children become 18, you know, you send them off to college and then you say it's their responsibility. Okay, maybe I shouldn't say that in front of a bunch of Chinese people, but that's that's how we do it in America. You know, my mother said, >> "Get out. never want to hear from no not quite never want to hear from you again but like now everything is your responsibility I think we should relate this way to our ideas yeah so um >> yes >> uh yeah maybe that's all the time I have >> one minute left yeah so I think it's fine to take inspiration from cognates I'll leave it there Yeah. Yeah.