 Hello, my name is Chris Foster. Thank you so much to the Hanumanology project for having me here. I'm something of an interloper. I'm not a linguist. I'm interested in linguistics as we'll see obviously. I'm also not going to be presenting on graph or network theory. Also something I'm interested in that so as in London I'm working on medieval Buddhist glossaries and I'm interested in the relationship of citation practices so I do want to do something with network theory. But more in terms of within a gloss how are different reference works related to one another and why. So that's kind of where I'm coming in from that angle but my first love has been reading these newly discovered excavated manuscripts. In the Han period and a little bit earlier written on wooden bamboo. And that's where I get into linguistics because I'm really interested in okay how can we reconstruct pronunciation of these old words and how does that then inform us in terms of reading the manuscript itself. So when I get into this, what I'm interested in doing is then thinking about what are these excavated manuscripts. Tell us about historical reconstructions. How can we use this new data set for linguistics. And so basically what I see here is an opportunity, and that a lot of the historical reconstructions for old Chinese in particular have been based on one specific texts, and all these texts that have been transmitted over time through the centuries that have been subjected to editorial and interventions. But here we have this new massive new corpus of data that we can then test that against right and find problems. So basically that's what I want to do I want to take sort of theories that have been developed from looking at the transmitted purpose the receipt purpose, test it on excavated manuscripts. And in particular, I've picked in terms of the excavated manuscripts scribal primers. I do this for a number of reasons primers. This is a site where learning to read and write first happened. It's part of education, it's part of a standardization process that was happening during the Han dynasty. It is also a fairly robust corpus so here I'm looking at one particular text called the subject pin. We now have it in I believe 18 different caches you can see it from all across China. You know, East Coast Northwest into central and the Southwest. It's a good geographic range. And what's really interesting too is that we can take these texts, and we can use them to anchor language usage in a specific time and place, right. We know, maybe not in terms of the composition of the text itself but in terms of one reception event. We can date this one manuscript to Gansu from the first century BC, which is really helpful when we're testing these, especially how does language change over time and in place right. Another aspect of this corpus that I think is really useful is that these primers tend to be based on structural rhymes. So the organizing logic behind what are ultimately just a list of random terms isn't so much the content there is sort of clustering of somatic relationships, but it's rhyming. In particular with the target can each chapter will participate in a single rhyme scheme, you'll see that there's four character lines, there's a rhyme, always after the second every second line right. And so what this does is it takes away some of that guesswork of, is this a rhyming relationship or is it not. This could be a problem and is when we get into the data, you can challenge me on this but as sort of my base assumption, I'm going to prioritize the structural rhymes over, you know, reconstructions. So, what, what I wanted to test in particular, in terms of picking just a random theory has been this theory of about old Chinese having a final R. That then changing into final N or final J. So this is forwarded by Baxter and cigar. They are adopting the theory from star Austin. Basically, but where this comes from is that there's been the observation that certain words have both middle Chinese final N and final J readings right and so how does that happen, or certain characters have this. We also see rhyme contacts in old Chinese between characters that have middle Chinese pronunciations of N and J so at some point these split off in a different way. And so in the hypothesis is that oh well way back in time and old Chinese, they had a final R and then eventually they split off with the change from R to N being a mainstream change, and then the change from R to J being something that happens in dialect. There's an idea that maybe this is East Coast Shandong, and that, although that wasn't adopted into the mainstream, some of this survives into middle Chinese. Okay. So now, another apology in that this talk isn't really sort of a polished paper I don't have conclusions to give you really I have some suggested conclusions. But really what I wanted to do is just go in a rough sort of chronological order from latest into older primers. I was going to show you the data and just give you a few thoughts about what the data is like and actually really use this as more of as an actual workshop, and get your your sort of helping your thoughts on it. So the first text that I was going to look at is called the G Joe PN. This is the only text of the bunch that has survived into the received corpus. Mainly actually through calligraphy it was used as a calligraphy model from some very famous players like one teacher. So we have versions of this in our received corpus. We also have manuscript discoveries. So there's a few here we have a wooden prism that dates to the Han period we have another paper version of paper manuscript witness that's slightly later. So here's a lot of the rhyming that I'm going to show you is based on the received text I'm treating it as late late stuff right so it's potentially Han dynasty but there's potentially editorial interventions as well. So here's some of the data. Basically what I see when I see this is an inconclusive about what the status of our is in this text. So what you see here in the highlighted in blue is a rhyme series that has been proposed by a lot of pay and jokes more. The G Joe PN isn't as strict in its structural rhymes as the science you can so we see that there are, you know, chapters that are wholly based on a single rhyme series but it's not inevitably the case. So there could be this this sort of grouping could be wrong. But again, I'm going to start with that as my sort of baseline. It appears here that. Well there's two interpretations of the material that one is that our has already changed the end. And the reason I say that is because we have towards the end, we have this yen which has a definitive and its old Chinese reconstruction. We have words like Chen. See here. Yeah, so yen, and then we have Chen here which has a definitive are in its old Chinese reconstruction and these reconstructions by the way are those that are posed by Baxter and cigar with some fairly mechanical modifications to update some of the changes that have been proposed as pointed out earlier. Right. So one reading of all this is that okay we have, you know, are is rhyming with n here. And so therefore, there must be a change. Sorry. So it appears that our has changed the end. However, if you'll see here there's a lot of sort of indefinite. Propositions for the finals right it's this maybe and it may not be maybe and so forth. So if you look at sort of weird stuff is being clumped together. It's possible that there is a smaller sort of subseries here where our is actually separated from and then that would suggest that our hasn't changed it right so it's inconclusive. So like, you know, here there is a little bit of that going on. There's other issues with this data sets like I don't know why, you know, here we have been and there's the ball E, whereas everything else has me ball of a. But again, I'm, I'm prioritizing sort of the rhyme series over reconstructions and so it's not going to bother me so much. Here's another series from the djpn. So this this reflects a little bit better in terms of what I'm saying about subseries. So we have our final our rhyming with final and one way of interpreting all of this is that okay, our has already changed the end. But here it's interesting that we have sort of three words that potentially have our kind of grouped together. And then we have the words with and being grouped together towards the end. Right, so is there a divide here is these two are these two different rhyming series versus one longer rhyming series and I think if, you know, if it was one longer, if our had already changed and this is indeed one long rhyming series and you start to see our popping up in more isolated incidents that makes sense. Yeah. Oh, sorry. Yeah, this is first century BCE in terms of so the traditional association is that this was composed by a fellow named shurio and 40 BC. However, I'm not treating it as late Western BC weight Western Han because we're using a transmitted version so my basically what I'm thinking about this is like this is sort of said to be on text, but because we're using data that's passed through so many hands. There could be later interventions and so I mean I think what really like excites me about the Han phenology project is that a lot of these attributions for supposedly context are incorrect and lead right so like a lot of this poetry and poetry productions are medieval. And what I want to be able to do is be able to go back and do exactly what you were doing before and say okay let's guess where this comes from and be like oh this isn't actually on this is much later. So I'm kind of treating this text in that frame that this is could be a context, we have an attribution have an idea but we're not so sure. There's a couple other things that really bother me about this series in particular one is this sure here, I have no idea why that's in the series or what could be going on. There's a whole long little bit of sort of hypothesis here that I don't think we really need to get into. But also I find that in a lot of these series we get Ren, which I changed here to a final end but in the old Chinese is a voice Veller nasal then. And you see a lot of voice Veller nasals appearing in these series and I don't know what's going on there and why that's the case. So you can see that here as well. Yeah, so here's Ren again with Jun, so that that string in now with our here we might have a smaller sub series there's a definitive are definitive are and then an unknown could be an end but it could be an R to so is this an isolated sort of smaller rhyme within other rhymes that are going on or is all one long series. Yeah, it's middle Chinese factor middle Chinese. Yeah sorry so old Chinese on the left, the rhyme word and then this is middle Chinese in terms of leon. Yeah, so thank you for sending that I just haven't had a time to put that, yeah, put that in but Oh, yeah, yeah, exactly. Yeah. Yeah, for Schuessler the leon. Yeah, yeah. No, thank you for that. And so another issue speaking about okay are there smaller subseries of rhyming here is the vowels as well. So we have a lot of schwaise, but then towards the end here it starts to get into, you know, I is a lot of eyes as the main ball ball so what's going on there. In general I've been basing a lot of this on just looking at the Dakota the final and not including the main ball there even though I strictly speaking it should be involved plus Kota but that does not seem to be the case often, even with my most strictly structured primers. Yeah, there's anything to hear. Okay. So let's go back to the time to a pen so the time to a pen. There is an imagined attribution to it as well going back to the chin dynasty. Supposedly famous figures like Lisa and Jock out had their hand in this I think it's mostly hogwash I think this is a story that was imagined late Western Han, when they were ridiculing the chin and they wanted to, you know, paint these figures and a negative pen as I mentioned before each chapter has its own one rhyme series, it's every sort of eighth character every two four character lines. What we have here I'm not going to start with the base tax to that but what we have here is a manuscript that was discovered fairly recently in 2009 out in Gansu in Northwest China. So it's the base text of the sanctioned PN, and then it adds at the very end of every line, three characters that start up their own rhyme series, which still is the same rhyme series throughout every chapter, but it's a novel rhyme series right. And this dates to the late sort of first century BC, right so this is our first sort of date and place anchor late first century BC. So here we have sort of a new rhyming series. And this is where we have our strongest sort of evidence of our sort of being retained in the text. So here we have Yin, which is rhyming with Jun, and then also she and I'm obviously using sort of modern pronunciations. But obviously so we have middle Chinese and middle Chinese ends with and middle Chinese ends with J. Right. So this would suggest to me that in this one manuscript for whoever compiled this commentary this on Japan. They did not differentiate between N and J yet it was R for these words. And what's also interesting about this is that because it's a commentary. It doesn't seem to have been very widely distributed or probably this is the only instance we have of this, my sense is that this is a very local composition and so this might help us get at whatever's going on in the Northwest right like this could be a vernacular in the dialogue. There is a problem, however, and that we have another series up here in the same manuscripts, where there is the incorporation of rent in what otherwise could be all ours. You know so inconclusive inclusive inclusive inclusive all the way down. The only thing we have here is Jun which is definitive are all of these could be final are. And then we have rent. Right, and it seems pretty clear that this word could not have been final. So how do we explain. Right. And I'm not sure I'm wondering it somehow it's anticipating the next rhyme, because the chapter after this then switches to voice Bella measles. Maybe we have sort of you know the final word of this rhyming group is actually not part of the series but then it just fits the next and memorization eight for okay now we need to move on to the next chapter. But that's a problem but I don't know how to resolve that. Okay, so now we're going on to the actual base text in such a pin. There's a number of witnesses to this. The one that I am going to pretty much solely use here is a what's called as the village teachers edition of the text and this one very strictly only does 60 60 words 60 characters per chapter. And it's best seen on what's called the Han board manuscript. Unfortunately, this is not an archaeologically excavated manuscript it was in a collector's collection. So we don't know, in terms of time and place anchors it's not as strong of evidence but we do have the same addition as village teachers edition, in other witnesses that have been archaeologically excavated. It tends to be somewhat later than the base taxes on Japan from pre village teachers editions. Which I'm not going to get into here. But again I'm thinking this is more like late or let's just say first century BC. I think the main the most important point to highlight here is this character couldn't where we see in what is otherwise a series with final Jay. There could be this word that has to end. It seems here that are. There was a final aren in old Chinese by this point for this linguistic community. This word had changed that final are to a Jay. Right. So this is probably the most exciting bit of data that I encountered with these primers. Unfortunately, this is only a tested on that one Han board manuscript so that's the big caveat, the big asterisk. So we don't have to be the case though if we can find and eventually get more data that shows that there was a change from our to Jay, or the base text of this on Japan. That'd be really super interesting and actually get us into questions about a where was this on Japan composed. We can start to sort of speculate about from an author who has connections to the east coast of China. And that gets us back into this sort of imagine the relation to Lisa and Joe Gao and all these other figures so there's a lot that I think could be done. I guess what I'm just trying to show how like these linguistic theories can then come and challenge these attributions and get us into sort of the social histories of these texts which I think is really fascinating. There is one of I don't know if you can really see the data that well but there's this character here. Right. Which, again, is this voiced Bella nasal, the in this whole J series. So this is another instance, I don't know why this is happening, but this just seems to be happening all the time with these programs. Here is a. Yeah, yeah. Yeah, so this is one of the only times that trying to think of if the other series of J series. Not as much as with M series there might be. I need to look at the, there's one more set the way lead your doubt types there may be an instance there but I don't think so. I think this might be the only time that's in J series. In other series where I think it ours retained it's still there. Even though those are then series that mean we seem to be words that change later to end so here is another primer. This one is a prism so that's one of these like longer sticks of woods that have four sides there's writing on each side of the wood. It was founded in 1939, also out in Gansu's Northwest China. It was found in the same sort of cash for these sort of refuse piles that had Sanjipian in it. And it has a lot of features that are similar to the Sanjipian in terms of the seems to be four words per line. There's rhyming it's 60 characters per chapter. The rhyme structure seems to be very different. So here we have the first three lines that's rhyming after every fourth character, and then it switches up. And I'm not exactly sure how to read this yet so it's not as strict in terms of the structural runs. But I've highlighted in blue the ones that I think are should be rhyming. And then I might want to include some of these as well, but I'm not so sure. Again, inconclusive evidence here are might have changed the end because we do. It might have changed the end because we do have a definitive are rhyming with the definitive end but it seems like the ours are only found towards the end of the rhyme series and so it's possible that maybe this is a different rhyme series than the ones about it right. So if that's the case then our has been specifically sort of, yeah, not changed. This is the last last group of primers called the way lead your doubt, sort of type of taxes, you know, so the way of being an official. These date mainly to the third century BCE so this getting towards the end of the chin. Not. This is before the Han phenology. And they've been found in central China who they. These are earliest finds. There are four witnesses that are extant one from D. There's another one in the academy collection. There's another that was archaeologically XBA I want to tie by a sense deteriorated there's been a lot of mold and it's not going to get published I don't think another one whose data we're waiting for that's in the peeking university collection. What I did here though is I mostly just looked at the, I've compared the straight to D with the yellow academy data in my paper that I've drafted for this. But right now I think I'm just going to show the straight to D materials. It's a much more complicated text. So, it's not built around strict structural rhymes at all. But there seems to be something close to it going on with, for instance, we have here like way. And then E and Dray like these seem to be something of a structural rhyme my color coded it over here. Right. That'd be those three. But there's a lot more sort of resonance that's going on in this passage. So based mainly on sort of Shua as a, as the vowel right so these are all main balls right all over the place. I guess the argument that I wanted to make with this is that if you look over this. It seems like it's read better if we assume that there's been a change from R to J. Right that just makes it more resonant. And if that's the case that's really interesting because that suggests that there was a very early change of our today. So what are what are sort of the my conclusions from looking at all this primer datas. So G Japan is possible that ours already changed the end based on the traditional division of rhyme series. But we can propose that there were further divisions within this longer series that retain a separation between our and so inconclusive with the commentary to this on Japan. There is very strong or at least very clear evidence of in one spot. It's sort of an R word that's rhyming with both and final and J file and suggest me that ours been preserved in that commentary. And that's really interesting because, especially for this commentary could be a locally composed and so that linguistic community retained are as a final. So if we look at the subject piano base text, however, there's some suggestion of an R or of an possible and final and a J rhyme series, which might suggest then that are had been changed to J already for the base text by linguistic community and an unknown location around the late first century BCE. And that conflicts that pointed out where we still have a rhyme that incorporated rent, which the voice fellow may as well and so I'm not sure how to handle that. My one module one, no significant addition, but then with a way lead your doubt it. Again, there seems to be a suggestion of an R to a Jane change that happened very early so in third century BC. That's basically what I've been thinking about and looking at. There's lots of individual issues that are in that data that kind of put in that last column so I'd love to talk about this some point and if everybody wants to look at the data I'll send it to you and get your advice on it. But my concern is mostly, you know, is there enough. There's enough data to actually say anything. It's still very limited and I did that on purpose because I want to be able to anchor these things in time place based on manuscripts but then I'm sacrificing sort of the robustness of my office. And again this sort of perennial sort of this enduring problem of how do we assign rhyming relationships and I tried to get around that by picking texts that have structural rhymes but even with the primers that's not always the case. And yeah, this presence of voice baller nasal in a lot of these series, especially the rent in the street and straight trends of manuscripts is bothering me. So, thank you for tolerating my talking. Yeah. And so you said base text and so this is not transmitted. Yeah. The only, so the only transmitted text from all these is the G Joe can that's on Japan. We have re compilations of this primer that we're done and like late imperial period. This is all the fad right. So those are based on our quotations of the subject and appear and other like reference works right so these Buddhist glossaries I'm looking at one of my the reasons I got interested in this because the subject and it's quoted all over in those that data, however, is very different because, you know, the subject can says that this word means this or rhymes with this or is written that way. But then when we found these new excavated manuscripts that for various reasons we can fairly confidently identify this was the subject can. Instead, it's just a listing of terms, you know, in these rhyming chapters right so clearly it's not the aren't glosses and the subject pain is saying this term is that or this term is written that way. So all of these quotations that are being compiled by these late imperial scholars seem to be later really medieval medieval interventions of interpreting the text. So it's useful data in a different way. One of them is the basic like you said okay. So really there's only one manuscripts where we see content that differs from all the others on Jim can manuscripts right in a significant I mean there's obviously there's individual variance here and why not. And if you compare the straight trends of manuscript to these other witnesses that we have, we're discovering that. So if you look at one chapter and the so called base text it's a four character line that you compare it to the straight trends that you have that same for character line, then all of a sudden there's three insertions of new characters, and then it goes on to the next four character line that we see in the base step, right. So it's mostly a comparison between the two that if we didn't have this straight sense of commentary there wouldn't be any. It would just be the subject. Right there wouldn't be a discussion of the base text. What's remarkable so I didn't talk about this like village pre village teachers versus village teachers edition. But there is remarkable stability across all of these manuscripts when I talked about two different additions. It's not that there's much content change between them. It's really just how they did chapters. The earlier ones tend to be much longer chapters 120 words together right, whereas the village teachers inevitably it's only 60 60 words per chapter. And I have a bunch of theories about that I think it's related to studying and studying on these prisms that can only hold a limited number of characters per. So, and I already had a question. The thing you just said makes it more urgent. So, the, each chapter has one rhyme. Basically, right. So then I guess on the one hand, you can have this 120 character. Let me try and use the pre use village teachers and have 120 character chapter. And then the village teachers have only 60 character chapters. Like, does that mean that the village teachers is just half as long a text and they cut out the second half of each chapter. They have more chapters, they have more chapters so basically you take 120 displays it too. Yeah, well but it's a little bit more complicated than that, because in the pre village teachers additions at least in some of the manuscripts we have to relate to that. It's not always in intervals of 60 characters right you can have 150 or whatever. And so there is a problem in terms of there was editorial intervention, when the village teachers made the tax where there's content loss or content shade. So this, this idea of like a whole chapter, having a rhyme, which I guess, sorry, rhyme, which I guess in the psychological literature is understood as is. Like, what text is that referring to those features. It's both, right it's the same it's the same content across both pre village and village. That's impossible, because I mean like, if you have, if you have 120 character. So, so there are chat it's not there aren't unique rhymes for every chapter one chapter participates in the single run, but you in the village teacher you can have multiple chapters that all have the same sort of ending. So it's not like they're trying to work. No, no. Yeah, yeah, okay. Do they have any theories about how this text is actually used, but what steps they like the kids memorize this thing and then write it out and play with chanted or something and then write it out characters. So this is kind of what drew me to study this in the first place is that if you look at this. The text itself begins by saying this is for young children to memorize, and if they do you can become a scribe. If you look at legal manuscripts on scribal training, while they don't mention this text by title, and I don't expect them to because I don't think the title was on this text until after these scribal statutes were created. It seems clear that this is participating in that same corpus. It's young adults, part of very, you know, it's an older it's, it's a, it's a, it's a career tool, basically it's not just, you know, I'm first learning to read and write and all these like little keys are. And if you look at the content to have later chapters especially I mean it's really sophisticated bizarre archaic vocabulary like you wouldn't expect your six year old to be studying this. And then if you look at like historical animals. We'll see parts where they're, you know, the historian is bragging about this emperor who, you know, could write the scribal treatise right and it's like why wouldn't emperor, like right about it, or you see a manuscript version of this buried with an aristocrat. This wealthy aristocrat like why would you, you know, I wouldn't bury myself with Dr. Seuss necessarily like. So there's a lot of different contradictions going on in terms of like, how was this text actually being who was it for and my sort of conclusion is that it was for these scribal families. It was used to test them and to make sure that they could enter into the bureaucracy that they were quick literacy in a certain way. These texts were very sort of strictly manage and controlled by those scribal families but that during the late Western Han it kind of gets out. Right. And I have a whole theory on this is sort of like the militarization behind frontier and he sends scribes out into the borders they got nothing else to do there's not strong controls. They start teaching it to the peasants and whatnot. Yeah, exactly. And so like a lot of what we see is from these military outposts and done wrong. And it's, you know, shavings of repetitive copying and trying this and it's usually the easier chapters the chapters that have vocabulary that's simple to learn. And that once that happens all of a sudden the prestige of this text drops off it's no longer used for testing. It's simplified which is becomes the GGO can see there's a lot of overlap in the content is actually there's discussions in the historical treatisees and say this was sort of made off of this on Japan, but it's shorter and it's easier. And that's the more prestige text becomes these confusion classes. So, and we have like 18 different caches that include this on Japan now the very first discovery that it was by all Stein in the early 20th century. And it wasn't it was identified shortly after that by one way. So, so, so there's a convention in early China where texts are often titled after their initial characters or characters in the first sense right so that's sort of the idea popped in a lot of these so there's another lots of discoveries in the 30s in the Northwest with Julian fragments of this text seem to appear in there, but really sort of like the big moment was in the 1970s. Yeah, 70s. Yeah, 77 right. And that was the first time we got one of these manuscripts in a tomb. That was a grave good. It's very fragmentary, but it was sort of a longer version and we can start to piece out the content a bit more.