 I'll just give you a sort of overall orientation into my thinking about the organization, this lecture, and I do this in part, so that you tell me what you've seen works in retrospect, is I went through all the imposters, right? And now I'm gonna go through the transposers, but actually I sort of cheated and in the beginning we use this whole thing in Christian series, which are transposers, but I sort of felt like, well, we have to, because some of you don't know Chinese characters, we needed to kind of look at how Chinese characters are structured, but then I feel a little bit conflicted, like maybe I should have just gone through all the imposters and ignored the Sheshan stuff and only done it now, yeah, so, because now I'm turning back to it and that feels a little bit messy to me, so you can let me know after, maybe we're at the end of the course, whether you think it should have all started now or whether maybe all of this should have been at the beginning, I just index that as a pedagogical question, I'm not really sure, so we've seen the Sheshan series before and now we're coming back to them. The first time we saw them, the point was just, there's such a thing as the Sheshan hypothesis, Chinese characters have a phonetic and a semantic component. Now we're looking at them, Kwa, the oldest available transposers, yeah, maybe not a particularly meaningful distinction, but what I have decided to do, well, let me just go to the slide, yeah, so reminding ourselves the Sheshan hypothesis, so only individuals in the same point of articulation are allowed in Sheshan series and then we, we'll do this in an extra fast way, but you remember, okay, we say the palavals are palavised dentals and we take the retroflexes back to either a medial R or a pre-initial R, so these are those proposals, then we have our lateral series, yeah, and then we have our voiceless resonance and then we have our more voiceless resonance and yeah, you remember all this, right? So I'm just refreshing your memory, these are all the simple consonants that we like and today we're gonna focus on its consonant plus three. And then last, the eugolipses, yeah, so you remember all this, anyhow, this, and then here you see the, this is, yeah, so this, this is the initial, we were talking about earlier, this is the special divided three voice, the other fricative, yeah, they use in their default reconstruction this voiced lego uvular stuff, yeah, whereas this one, you know, that's, oh, I can't remember, yeah, that's the other one, which should have had a little circle under it, oops, I really hit the wrong button there, so those of you who, you know, okay, here we are, back, and this is on the eugolipses, yeah, so that we all covered, okay, so, oh, and then, yeah, so now we move on to sheshan evidence for clusters, so when you did your homework, you ran into stuff you didn't know how to deal with, hopefully this will cover most of that, so summa becomes sub, and here you see that we have this character, which is just read as sum, so what is it doing in this series that's mostly read ma, yeah, and back from cigar's answer is it comes from sum, yeah, and then you will notice that in these cases, which I mean constant cluster evidence, I will sometimes get three chin forms, that's because in these cases, the sheshan relationship is non-obvious, but if you look at this character, you don't see this ma inside, but if you look at this character, you do see this ma inside, right? So is this summa script or what's final? So I'm leaving that to one side, basically, so I'm just giving you evidence that it's in the same sheshan series, and for that, I'm just giving you sum preaching form, yeah, and they're all copied from back from cigar's book, so they give sum more paleographic information, but actually not that much, and yeah, and if you like, the question is how old are sheshan series and when is each character in a sheshan series calling? Yeah, so that's an important philological question that there aren't convenient resources for me to check on, so in this case, I'm just whenever it's non-obvious from the computer font that you get, if you're writing someone an email, then I add a preaching form, yeah, but I think it's also not a coincidence that it's in these cases where the sound really veered quite far away from each other, that then the writing system has also not maintained the obviousness of the link in the sound, right? Okay, so anyhow, sma becomes ma, and then sma also becomes suh, sorry, sma becomes suh, suh, sma also becomes suh and pi b syllable, so here's an example on pi b syllable, yeah, so this split is the, is the, is the phonetic in these other two characters and then here you get, it's nice that here you get the voice with resonant and you get the sm cluster and then you see why they need both, right? And then, you know, it won't shock you, when sm becomes s and sm becomes s also in type b syllables, sma becomes suh and then here's where the, I think we looked at this series before, right? We looked at this series and we saw that Carlvren sees this as its own shape and series, as the head of its own shape and series, but back from the cigar, see them all as the same shape and series and you can understand that as being Carlvren had no hypothesis at his disposal for explaining why on earth this character would have the fish in it. So he said, oh, it's its own shape and series, yeah, and it starts with an s, whereas back from the cigar, say, oh no, it goes back to a sma, so it makes sense that it has this fish in it, yeah. And also in type b syllables, yeah, okay. And then sla becomes suh, yeah, so generally speaking, s resonant clusters just are reduced to s. Now, is that a reasonable sound chain? It actually happens in a Tibetan dialect. So like for instance, in a particularly Tibetan dialect, the written Tibetan word no snuh just becomes suh, yeah. So it is a sound change that I can parallel. Yeah, and then this one's nice too, because why is in the lateral series, well, you have the yuh, the connections, and then you also have this thuh. So now you see, we have this extremely messy sheshan series, but we have some tools to clean it all up to being lateral initials, right? Yeah, so, yeah, t could not hurt. No, t could not hurt. Yeah, and actually that's a good way of putting it, because it's not like we can just do anything, yeah. Or our hypotheses are carefully calibrated to only give us what we want, it's maybe the right way to put it. So you get these and th's and y's and s's, that's a lateral series, but you won't get a plain t in such a series. Yeah, I'm just tempted to say as an aside, so all Chinese does have t, th, and d, but there aren't very many th's, and I mean this for all places of articulation, in all Chinese, and then you see like, a lot of the th's and middle Chinese are coming from things like laterals and various clusters and whatnot. So, I'm not that it would be a particularly good assignment, but looking at, is there some way to get rid of the remaining aspirates in all Chinese, it's a, is it right? I'm even disappointed, yeah. I should probably say that next week, but probably we haven't been there. Okay, and so then we also have s before q, so here's a, and whenever you have the voice, let's be affirmative, but that's indicative of a uvular series, because now we have an s coming up in a uvular series, so then we take it to a cluster s plus uvular, and then also in type b syllables. Okay, and then why not with the h, but here I say, it is not clear to me why bad students cigar choose sqh, rather than sq, yeah, they won't say anything about that, yeah, and yeah, I don't, there's nothing that I know of that would motivate that distinction, but they make it, yeah, and what do I say? And we could also take it back to a uvular series, right? So I'm not sure I've discussed this thing, but in their system, muh before, I think I did, muh before a uvular, it gives us a vular nasal So if you see a vular nasal, an h and an s, yeah, you could take it back for a uvular, but you can also take it back to a vular nasal by having just a vular nasal come from a vular nasal, the edge come from an s in front of a vular nasal, and the h coming from a voiceless vular nasal. So this is the sort of place where like, I don't know, where maybe we have slightly too much machinery available to us, or at least it would be totally as far as I like to tell, in keeping with back from to our system to reconstruct this way, but this is not what they reconstruct, they reconstruct this way, and presumably they have reasons too, but they don't always say what those reasons are, yeah. Okay, and then here things are still pretty tidy, so S, T, S goes to S, and if we have a sheshen series that mixes T, S and S, then maybe we can take it back to an STS, this is a sound chain that actually happens in standard events, STS becomes S, and here we can tie these symbols, yeah. Now, SR is going to become a middle Chinese, a red reflex S, which you would expect, right? And here we have, you know, so we have L's go back to R's, so then a K, and then a K in such a series, we're back to AR, I'm not sure that I discussed that, although it did come up in some homework, and then what do we do about an S in that series? Well, we can make it SR. Now, they, well, let's see if I understand. No, I don't, yeah, maybe. But they end up having S, like SR with the R being the initial, and SR with the R being the medial, as having potentially different outcomes, and I don't think that's great. Okay, so, and then now, basically this is just, I'm going through it again with the R medials because this red reflex R is a different outcome in middle Chinese, according to middle Chinese phenomenon, but it's all what you expect, right? So, we have, you know, that, and then we can also get a palatable S's. So here we have these palatable S's coming up in what otherwise would be a dental series. So why not take it back to an ST, Gintype B? Just going to, you know, okay. Now, here we have a K series, and we have a shot in it. So why not take it back to SK? Yeah. Sorry, the R series, okay, well, which comes out as L, then it can combine with K, but look with G or other. No, it can, I just kind of, like this was just an oversight in, you know, three days ago or something, which is if you get an L series mixed with, is it just with dealers? I think it's just with dealers. Yeah, then you, I mean, it's still a sort of tricky question for the sheshen hypothesis because, you know, which way it should start, kind of thing, your initial work, yeah. But there are, and actually maybe another way you could, you could sort of discuss, you know, if you want to go back to this, where are the R's? Are they medial R's or are they pre-medial R's? Some series have, let's say, have dealers where sometimes the vowel coloring or other things make you want to see some kind of ariness in them, but they don't have actual characters written with love initial, yeah. So there are those where you might say the retroflection or it's not a retroflection, the vowel rewriticizing comes from the initial sort of pre-initial, but I would say if you actually have R initials, old Chinese R initials, the middle Chinese initial L's, as full-fledged characters in this series, then that means it's definitely gonna be an R medium, yeah. And what I would do actually, let's say, just if we want to be sort of mechanical, and I think at one point in the past, we had a vaccine cigar, didn't we? So we have a series where we have something like Chiang and Lang in it, right? And then maybe I'll just say, we also have that word. Well, I don't know whether this would actually happen, but let's say it does. So just given, so you know, the E-writic is generally go back to G, so then, and L's generally go back to R, and then K's generally go back to K's. So then we have K, R, and G initiation series. So then you want to say, oh, somehow the, to use the term I used earlier, the Somus Gramai, I shouldn't put the star, of this series is Kram, right? And then we would, if we were romanizing, let's say we have, you know, this one has a tree, and then I'll just add symbol, and then this one has water and add symbol. You know, we could say this one is R Kram, and this one is, and then this one is Ak Kram, right? But then you say, yeah, sure, but, you know, R and K don't have the same place of articulation, right? There's still a problem here, right? And a mechanical solution that I believe at one point in the sort of early 2000s, back from cigar days, they would take this one back to, or here I'll erase them and write the proper reconstruction. We take this one back to Kram, and we take this one back to Kram, and then we would take this one back to Kram, and then that way, you know, you're still getting the velar initial here, right? But the Kreefix would drop out. So this solution would work in their current system, like Kram would change it the long and middle Chinese, and then this series would get you conformity with the Sheshan, Sheshan of others, but it turns out they generally don't reconstruct this anymore, which is they don't see long or long coming up with Kram as sufficient reason for reconstructing a velar pre-initiative. And why don't they see that? Maybe it's because they don't get the other evidence for the pre-initial from loans that they want or something like that. They don't really talk about it, yeah? But I would say that just if we are sort of naive about it, this is a great solution for these LK mixing series. So we're not even looking at such a series, but we did have one, but where did it go? Where did it go? Did we not? Where did we not have? Oh, no, this one. Yeah, this one. Right. So yeah, so like if it were me, I think, I'm not actually saying that, I'm just saying like if you were driven entirely by the Sheshan hypothesis, I would reconstruct these first two with this kind of Kram or something like that. But then what would we do here? Would we do a Sukran? I'm not sure. Or maybe we would do Kosran, yeah? That would be better because this wouldn't get what we want in between you. But then of course, we're sort of violating the Sheshan series if I call this as anyhow. And actually maybe this kind of difficulty that we run into in a series like this is why they don't throw in these Vila prefixes to make these ones look better. Hi, can you hear me? Yeah. So if I remember correctly, I think they still like reconstruct a tightly attached K prefix for the Sheshan series involving the word for ghost, like for the Gloucester series. Yeah, yeah, yeah, yeah. Thank you. So I guess this is a question I've been wanting to ask is, I mean, if they have this strategy to like repair this Sheshan series by involving a K prefix here, I kind of don't know why they still do the Uvula thing. I mean, certainly in this series you're talking about it's a really good question, right? Would you say if you, let me just actually explain to other people, which is if I say, okay, so let's just imagine a Uvula series. So I have, oops, I should try and write in my system. So I have some of these, some of these, maybe one of these, yeah, and then a K, right? So given what I've told you so far, you would say, okay, this is clearly a Uvula series. And this one we take back to a QH, QH, this one we take back to a count of G, this one we take back to a Q, and then this one we take back to some prefix before solving Uvula, yeah, like this. Yeah, something like that. That's what they generally, I like to do like this, right? That's what they, that's what I taught you, yeah? But what, actually, who asked this question? You're just gonna get here. Hi. So the question is, they can also get this K with K prefix before the long stop. They allow themselves that. And then why not say, this is this, this is this, and now we actually have the answer because we won't be able to get these from the Vlado stock, yeah? But in a series that only makes Vlado and K, there are two options, yeah? And as far as I can tell, I mean, I think we either did this or we'll get to it. They only introduced these two options because otherwise they end up with two sheshen series but indexing the same syllable type, which is something they don't want to do. They like to imagine that each sheshen series points uniquely to a syllable type. Oh, then this might also be the reason they reconstruct, like the one of the series we mentioned earlier with the Uvula instead of with the Vlado because there is, because the series for five is with the Vlado. Yeah. Yeah. Yeah. Does everyone understand that? I wasn't here. Yeah. So I was working for this. And then, and then I was just sort of tricked into giving the answer, which is probably because this is what they reconstruct for the Vlado stock. Yeah. So this is what they reconstruct for the series with the five. Yeah. So this. So this goes back to now. So if they, if they have, if they reconstruct a series based on none, then they don't want this one to be none. They want this one to be something else. So they do this with it. Yeah. Yeah. Yeah. So the argument here is that move off that will not do me the So I'm just going to say we have the same series here, although it doesn't exist in quite this way. Right. So that would be very, very easy. I agree with you. This one would be. So then, you know, imagine that we have this as our series. So they would reconstruct this one like this. Like just like I've said here, right. And then like this. So let's say on Monday, they reconstruct these. And then on Tuesday, they want to cross this series. They say, oh, we don't want to reconstruct it like this, because then we would end up with two series reconstruct. For the same syllable type. So then they see what other, you know, what other ways they're tools they've allowed themselves to get them the right instructions. And then maybe they do this. And so that's the hypothesis that they, I say, well, why don't we do the similar thing? Maybe the answer is because for these ones, they've already used up that sort of spot. Yeah. Now, of course, they're still a question of like, if you have two things and they could both be written and start the two ways, I've decided which one to do which way, right? Although I would say in this case, it's been easy because the Tibetan languages are pointing unambiguously, unambiguously to reveal your nasal. But this is the sort of thing that they don't make sufficiently explicit. Now, I think, you know, they would say, which would be very helpful, like, look, it took us 10 years to write and just like for on basis, right? So what do you want? But, but, you know, I always want more. You get the impression that the, yeah, all of these distinctions are inspired by go across someone that they need to account for. But as a result, they, you know, they can account for more than there actually is. Yes, I think that. Yeah. Well, and you can, especially if you get like, for instance, you know, you get like, say, any team. And we're now, let me say, let's say how many sources of D are there. Yeah. Well, actually, let's say how many sources of. Yeah. So one is. And that's the most obvious one, right? But then you could have. You could have, you know, you could of course have. And then you could have a lateral. And then you could have, if it's before a front valve, you could have a G. Yeah. So, you know, you, it starts to feel like. Sort of. There are no rules, right? You can kind of reconstruct whatever you want. But I think that's not quite true, right? Like, like they, they, they do set out rules. But the other question is, do they always follow their rules? My question is they generally do, but there are a lot of rules to follow. So it can be hard to know, right? But I think the, let's say, I think the way, let's, let's all wax slightly philosophical here and say, this is exactly the approach by in terms of reconstructing lots of distinctions that Carlton first did, where, where he basically took all the distinction in middle Chinese and back rejected them on all Chinese and then added more distinctions in order to account for the connections that you see in all Chinese. So you ended up with like, you know, 50 vowels in all Chinese, something like that, right? So it was ridiculous. But then, like people like O-Chang Pei and, and, and pulling blank just came along and said, actually, you know, this value reconstruct only happens after dentals. Right. So they were able to clean up his reconstructions very effectively. And then I think the question is, are there gaps, are there complementary distributions in Baxter in regards to all Chinese reconstructions? And I don't remember exactly, but I have definitely found a gap. I don't remember what it was, but I think this is the sort of thing that needs more exploration is just, you know, look at, you know, maybe, maybe they, they've given themselves too many tools. So now they can explain far more than actually exists. So then if that's the case, it's worth asking what, what syllables do they allow themselves to reconstruct, that they never in fact reconstruct. And then, and then that might be a way that we can kind of simplify the reconstructions. Yeah. And we're doing, you know, we first, we, we play S, then we explain strud, then we spin shop. Now we have Z. So it's a Z pop-up in a Uvular series. Then we take it back to an S before. Right. Now, interesting to note that, right, we didn't see the S voicing before an M, before an L, before lots of voicing things, but we do see it voicing before this Uvular. And then I say, okay, but imagine, so are also reconstruct S, B and SMT as sources of Z, but not on the basis of Shashank series. They do it on, with other elements, more volatile elements. Okay. And then we can also get this retroflex Z in this series. Okay. And then I just mentioned this, which is usually, we, well, yeah, there's an early palavation of V-load before front valves. And so then we get, you know, this, this job when we usually take back to a D, but here it has a connection with the G. And then we also have the Z in this series. So we reconstruct this series this way, but this would, this is a change. It only happens before front valve. Yeah. And yeah, so I just sort of said, but so in general, they think SG develops into G, but they're having a, so this is not based on Shashank series, but before front valves, SG develops to Z. And I do actually would say the fact that there are these, you know, sound changes that only happen before a front valve does kind of, or let's say either support some of their reconstructions or you can say your reconstructions are designed to deal with it. Okay. And then now we start to get slightly, if you like, stranger proposals. So far, S clusters have gone to S, they've gone to retroflex S, they've gone to parallel S, then they've gone to Z, they've gone to parallel Z, and they've gone to retroflex Z. But here we have kind of changing into Shashank, yeah, which feels a little ad hoc, although it doesn't, is it's not a crazy proposal, but feels a little asymmetrical, right. And then this is the evidence for it. Here you have a global stop and here you have a stop and, you know, global stops either have to go back to global stops or they have to go back to uvulars. And we don't have any way of getting what we want here. So we propose a new sound chip. Okay. And then same thing before a, before an R, it's just a retroflex in type, no, sorry, in type B circles. So this one can have a global stop. Okay. So you can literally say it's something they have to use more than once. Okay. And then a very exciting one, I think, is S before a snub. So snub, they take as going to suck, right. So that also seems a little ad hoc, but it's great in terms of explaining the relationship between these two characters. And, and everyone agrees that the word for a thousand starts as being the word for person with a line through it. So it does seem like, you know, that's, I mean, this goes back to me. Yeah. So this one goes, this one name, and this one name. And, and we can't, you would say, actually, like, well, so, so, this one is the total run is, is mean. And then this one, you can't take it to sing, because that would become seen, right, regarding, regarding talking about that. So instead they take it to mean. And then they say there, this sound chain. The s and the m will be one. Sorry, s what happened to s? S n goes to m. And we have not discussed s. No, we haven't. That is not a reason. And that's actually, you know, the sort of thing where like that. You know, you could ask yourself, well, why, why aren't there any smuzz. So, so let's just keep trying when we end up with a snuff becomes. So, and here it's love. So we are starting to see, you know, that it's not totally at all. Right. So here in this case, classic B and yell means we have a lateral, but then we have this song. Well, why not propose an S cluster. That's, I think it for the most reason, but you say, we don't have smart, right. And then you would say, is that an accidental gap or is there something wrong with the reconstructions. Yeah. But then, yeah, here we have, they have s T becomes so. And this one sort of makes sense because we get the same thing, but wrote it, right. Or yeah, reflects, right. Okay. And then we're going to the really exotic things now. They see as a source for links between and this is the one example as far as I know. So they, yeah, so they think it goes. So that's their explanation. Okay, and then I think I'll stop there because that was everything you can do with S prefixes. And that's most of it. But then we will have some other prejudices on peas and some cheese. Okay. So any questions. How many percent of the children's years do you think work with just the place. I have no idea. I would say, basically, a lot of it has to do with how long they are. So I would say, maybe half or more, you can get with the sort of without the clusters, but any really long series is going to have something funky in it. That's my sense, you know, a lot of patients, there's only three characters. So you can get. Yeah, so you don't need to, you don't need to be done for them. So we think that's more than. Oh, yeah, that's a good, I mean, if it's all starting to feel a little too incredible to you. I suggest just stare at each other in series. And then you say, there is something going on, right? Like, I think it's because, because most series have like, you know, let's say, I don't know, let's say it's like our slow example. Yeah, so, so you'll have like the why, why, bro. And then I don't even know what this goes to. So you have a lot where like the, the normal hypothesis, the sort of without the big guns explains almost everything in a Shashank series, but then you have these, you know, what are nuts to crack, but, but you don't want to say, oh, you know, because what are you going to say, well, they didn't know where to put this. Why did they put this one in a series, right? Like, like, you need some explanation of what's going on there. Correspondence of the latest two particles, something similar also happens in iron. So some iron dialysis for a while other iron dialysis chair. You're talking about what where they do. Yeah, what's the TV. Well, how did the iron new all of this standalone. So there's one reconstruction that happens with. So programs, one dynamic and the other. And others just assume that, but that goes back to a part of life. So those are the two. So, so like a pun up here. Yeah. So it's event you have becomes, so it becomes job. So that seems quite, you know, and that is the sort of thing that I would rather some sort of, you know, be a be a posted job rather than. Spar goes to. Yeah. But I think the trouble is that we did this. That would cause other problems for them because because type visa was all paradise. And this is a very minor thing. Yeah. Okay. Are we sure that. Is it that it is possible to have the contrast of voices and voice mazes after this. That's a good question. Terrible. That's terrible. And you'll be happy with that. I see the reconstruction from having a book about. I don't know of anyone commenting on the type of logical thought video. Yeah. I will say that they, so they have to allow that, right. But let's say they think, so a lot of people want to reconstruct S clusters where they reconstruct the voiceless nasals. And they say when pressed about that, they say, yeah, sure. The voiceless nailers come from S prefixes, but there must have been sort of two waves of S prefixation where the sort of secondary S clusters became voiceless nasals. And then there were some secondary S clusters that, that change into S. Yeah. And so you can say that like if you, if you had an logical process of S prefixation and you had voiceless nasals, then you would at least have the occasion arise where you would want to put an S prefix in front of a voiceless nasal. Now, how that ends up getting realized maybe with an S prefixation? Yeah. In many cases. Yes. Yeah. Yeah. Well, anyhow, I'm going to like. Yeah. Yeah. Then that would then still make your voiceless nasals with the kind of S. Well, that's the question is, is, is to you feel like having a tight S before a voiceless nasal is, is something that a language will do. We can't say all S was able to say it's loose prefixes and broke it all, right? And that's what I really want to say. I mean, that's the sort of thing I think would be good to look into just for the, for the record. The question is like, again, are we giving ourselves too much machinery? So we have tight and we have loose Ss, and we have voice and voiceless nasal. So then you have like four different possibilities. This is the sort of thing where I think it's worth checking. Do they actually use all four possibilities? And if they don't, can those gas be explained by altering reconstruction? Yeah.