 So Chinese from 1250 BCE, which has a kind of logo, well, let's say a logographic script, but with a sort of syllable-based phonology. Pew, which you already saw, is kind of a segmental phonology based on a Brahmi script. Tibetan. Tongut, which is more Chinese style script, although it gives even less information about pronunciation. Burmese, Neuart, and Yi, and that gives you a sense of the accumulation of languages with the written tradition over time, each of which has some analysis of, let's say, segmental phonology and syllable structure. The next, I think, kind of circumstance where linguistic analysis comes up is diplomacy. If you want to, in the pre-modern world, if you want to talk to foreigners, it's probably because of a diplomatic or commercial setting. And as we saw, this was the case with Bailong. So the Bailong were, I mean, we only have the Chinese side of the story, but were a subject people, a kind of newly subject people to the Chinese state, and they presented these songs to the Chinese court that so impressed the emperor that he wanted them written down. So that's a kind of diplomatic context that has left us some evidence of the language. And the Huayi-Yi-Yu vocabularies, which I also mentioned, are similar. And actually they start with, they're a very complicated set of texts, and I wish someone would write like a kind of a handy companion to the Huayi-Yi-Yu vocabularies. But the earliest one is the secret history of the Mongols where the right after the fall of the Yuan, it was clear that let's say Mongolian was very important language still in China. And there's a text that's actually probably the earliest in a sense, Mongolian text we have, written in Chinese characters with a lot of special conventions for giving linguistic information. And that sort of started a trajectory of documenting foreign languages ostensibly with a diplomatic context, but not entirely. And it goes right up until the, basically the end of the Qing, and there are vocabularies on French and German and things like that. But just looking at languages that are interesting to us, Gyaorong is covered, Tosu is covered, several Tibetan dialects, particularly dialects of Sichuan, like Baima. I think Baima came up in the presentations of the students mentioning their interests. And Burmese, actually Vietnamese also like, and for Burmese, let's say for those languages with scripts, they give a script sample and the words in Chinese pronunciation. So this is a useful resource on historical phonology as well as, let's say in a sense, the very earliest generation of sign of Tibetan linguistics. So then we turn to self-conscious scholarship. I think the earliest is by colonial officers, British colonial officers and missionaries in Burma. And in fact, actually, one thing that is surprising is for instance, the Kuziqin languages. More research was being done on them in the 19th century probably than now. It's changing, but that's because when Burma was part of the UK, in a sense, it was very accessible to research, whereas then during the military dictatorship where there were international sanctions, it was less accessible. So colonial officers and missionaries and then increasingly university-based academics. But I do want to emphasize here that there's not hard and fast lines here. Basically colonial officials are gone, but missionaries are still, I would say, half of the work that's being done on documenting sign of Tibetan languages is still being done by missionaries. So missionaries remain very important and there are universities like Payap or Mahidol in Thailand that are associated with missionary work. So there's a crossover between universities and missionary work, but I also think that in the context of China where the government is governing minority peoples and running universities, there's also a kind of, if you like, overlap between colonial officials and university-based academics, particularly in the 1950s when the big ethnographic classification projects were happening. So now that's sort of my bird's-eye view of how we get from diplomatic contacts between the Han Empire and the Bailong to today. And then I just look at sort of the orientation that linguistic researchers working on sign of Tibetan languages tend to have. So field workers tend to have a functional typological orientation. And I'm not gonna talk a lot about what that means partly because I don't know, like, but it's a thing people say about themselves. Is that they're functionalists and typologists. And I think basically that means that they're not Chomskists, but maybe it means something more precise than that. It does mean that people are interested in grammar and in some broad sense in linguistic theory. So we study funny little languages in Asia to contribute to the understanding of language. And that means lexicography and text collections are undervalued. And I'll say, maybe indicative of my training as a philologist and in Buddhist studies, this is something that kind of grieves me. We don't study Buddhist texts from the Tang dynasty, for instance, because we're interested in contributing to linguistic theory. It has to do with trying to understand those people and their time and their culture and a kind of rich diverse set of human experiences. And unfortunately, ethnographic work on sign of Tibetan languages tends to not give attention to those areas. And I also think there's a sort of paradox where a language that's easy to document is not easy to use in comparative work. And I mean that in a number of ways, like I think that there's a general tendency, I mean, this will sound a little bit outrageous, but I think it's pretty established that languages in kind of warm, flat places where lots of cultures have come into contact tend to be fairly agglutinative and have relatively simple phonology, whereas as languages spoken in tiny mountain villages that are very isolated have more interesting agreement systems and more constant clusters and whatnot. So that means that a language that's hard to get to is going to also be hard to document in terms of its actual structure and a language it's easy to get to where they will have comfortable, delicious food, will have a more simple structure and be easier to document. So I do think that there's a pressure there for documentary work to be focused on a certain kind of profile of a language. But unfortunately, for historical purposes, the more strange and complicated language the more useful it is because just the more machinery there is to study its history. Whereas if a language has a quite simple structure in terms of its syntax and its phonology, that means a lot of information has been lost. So I think there's a tension there that has meant that the languages that have been most documented are not as useful for historical linguistics. And like I said, even among, let's say, among everyone working in the field, there's a lack of attention to philology. And I think this is because Sign of Tibetan, historical linguistics interact so much with field work and not with text-based scholarship. So that's my sort of, yes, the overall profile of researchers in Sign of Tibetan linguistics that has an impact on the kind of work that can be done and the kind of work that is being done. And now I sort of discuss the various scenes. And I don't want this to become too scurrilous. And maybe it's good that we can't go out to the pub or anything like that, but I do think it's useful to have a bit of a sense of the kind of the different scenes, the different, the sociology of the discipline. So kind of schools, if you like. So there's what we can call the Berkeley School, which was associated in the 1930s with a big project as part of the kind of New Deal era to amass data on Sign of Tibetan languages and write something. They wrote a big 12 volume book that unfortunately was never published. There were three copies made and two of them are in the Berkeley library. Schaefer left the project in the 40s. Benedict took over and then Benedict wrote the book that was eventually published in 1972. It's been very influential. And Benedict left Berkeley and moved back to New York. And it was actually, and then Matasov did his PhD at Berkeley and had his first job at Columbia and met Benedict there where Benedict was working as a psychologist. And that's actually where then the continuity happened in terms of the, you would say the parampara, the lineage, yeah. And then Matasov himself ended up working back at Berkeley. So there is this long tradition of Sign of Tibetan linguistics at Berkeley from the 1930s until Matasov's retirement around 2015. Now there's no one at Berkeley as far as I know, but Matasov I think is sort of, I don't know what he would say, but let's say the student of Matasov's who most presents himself as heir to this intellectual tradition is Randy Lapola who is at Nanyang Technological University. So in Japan, there's less, maybe of a definitive kind of lineage, but I do think that it's important to just mention that the Japanese are very strong in Sign of Tibetan linguistics and particularly Tatsu Nishida shines as a sort of luminary who is someone who did fieldwork on a bunch of Lolo Burmese languages and was perfectly competent as a tibetologist who wrote a monograph on Burmese as preserved in the Hawaii vocabulary. And all of that is a tiny part of his research, most of which concentrated on Tangut and the fact that we can kind of read Tangut with relative ease has a lot to do with Nishida's scholarship. In a way, it's unfortunate that he didn't kind of found a school, but there are lots of people still active in Japan, largely or let's say interacting closely with the Berkeley school, if you like. But then the other person I want to sort of name is Yoshio Nishi who worked very closely with Nishida on this big encyclopedia, the Gengogaku Daiji Ten, which is the big linguistics dictionary. And it covers, it's this sort of reference work on all of linguistics, but compared to similar reference works in the West has a lot more on sign or Tibetan. There are 40 different languages covered just by Nishida and Nishi with articles of sort of 20, 25 pages. So I think that's a very useful resource it doesn't get looked at as much as it should. And then Nishi himself was not nearly as productive as Nishida, but in my experience, everything he wrote is rock solid, like really insightful. So he's another person who I think is worth mentioning. And in my earlier presentation, you noticed that there's still lots of people active in Japan, both based around Osaka and in Tokyo University of Foreign Studies, but I'm not going to go into any details there. Then there's what I always used to call the Leiden School, but then as I wrote that on the slide, I thought, well, now probably the Leiden-Bern School associated with George van Dream. And I would say the hallmark of people associated with that scene is very ambitious descriptive grammars for the PhD. So I think that, how can I say, a surprising amount of the new knowledge about signative and languages has been published by students of van Dream in their thick language volumes, most of them published by Brill, but not all. And that's a different scene that has been more focused on documentation, a little less on comparative linguistics, but I think that's all for the best because you need good data to do good comparative research. And then there's a scene that we can call the Oregon scene, sort of built around, if you like, Scott Delancey, but also ideologically heavily influenced by this fellow, Talmi Givon, who is not a researcher in signative and linguistics, but is a sort of theorist who is also at Oregon. So at Oregon you had sort of Givon as the theorist and Scott Delancey as a signative and specialist. And he's also had a series of PhD students who have done documentary work on signative and languages. And I think his first, she was PhD student, Carol Janetti, is herself just retiring this year, has been based at the University of Santa Barbara. So there's a sort of, maybe it would be better to call it the West Coast scene, except that that would exclude, because Berkeley is also on the West Coast. But anyhow, so that's another scene. And then I think they have, it's a little bit artificial to say these things, but had been really prioritizing typology per se. So whereas Matta Saw students were doing reconstruction and Fendrim students were just doing sort of very traditional descriptive linguistics in a more or less structuralist approach. In Oregon, they focus a lot on typological characteristics in their descriptions. And then there's Paris, which is kind of newer to the scene of scenes. And I would associate with Guillaume Jacques, who is at the CNRS in Paris. And he works on Ghearon and his students mostly work on other, let's say, other forms of Ghearon. And that has created a kind of tight knit group that is writing research that interacts with each other very profitably. And as I mentioned before, I think one thing that Guillaume can be commended for is just noticing the importance of the Ghearonic languages. And their documentation has benefited Sino-Tibetan linguistics a great deal. There's something like 300, 400 Sino-Tibetan languages. So we're clearly not at the point where we can reconstruct the proto language and explain how it emerges as different daughter languages in the way that happens, for instance, in Indo-European. Instead, solid reconstruction is done at the subgroup level where what people do is associate correspondence patterns like this Latin, P, German, F with proto segments. And there's nothing wrong with that per se. In fact, it's in my own methodological remark. So a few hours ago, it's what I said should happen. But it tends to sort of stop there where you sort of find a correspondence pattern, propose a reconstruction and then reconstruct a vocabulary. Whereas in what you should do is formulate sound changes that state very precisely the phonological environment and the relative chronology of the sound changes. It's not enough to know that this change into this and this change into that, but rather you need to know what happened first, what happened next. And there is a tendency among reconstructions at the subgroup level to kind of sweep enough tricky things under the rug that any one reconstruction cannot be used to predict attested forms. And I see that as the kind of major failing of the reconstruction systems that are around. Is that in the best case, you should be able to say, okay, here's my reconstruction. Here's the historical phonology of this language and you can apply those rules to those forms and get the attested forms. And that tends to be impossible or nearly impossible using most of the reconstructions that have been published in Sino-Tibetan linguistics. So now this is back to a methodological principle about distinguishing inheritance and borrowing, which is why relative chronology is important. Namely that if two languages share a sound change that occurred in their respective histories subsequent to a sound change they didn't share, then it's a contact induced sound change or a coincidence. Which is, you know, let's say going back to our P changes to F example, if we could prove that, let's say English and German, let's say German was affected by some sound change that had to come before the P to F change, then we would have to conclude that the P to F change was a coincidence in German and in English, right? So let's say, I'm not sure I put that very well. When two languages share a change, the default assumption is they share a change because that change happened only once when they were the same language. That's just Occam's razor, right? It's easier to propose one change than two changes. But you can sometimes show that one of the two languages had to participate in a change before that shared change. I'll give some examples. And that would prove that they have to be two events. And we don't go through this work in general in signage Tibetan linguistics, which means we can't rigorously distinguish inherited vocabulary and borrowing. So just some examples of the sort of thing I'm talking about. Initial manner induced tonal splits around the same time in Chinese, Korean, Burmese, Loloish, Thai and Vietnamese. I realize I misspelled Vietnamese. So we know for Thai and Vietnamese is a coincidence because they, or let's say it's contact induced because they're not even related languages. But the question arises, for instance, when Burmese and Loloish underwent a similar tone split, is it one event because Lolo Burmese was still one language or is it through contact? That kind of thing has yet to be well worked out. Another example is there's a change of S to T, which seems quite strange. It's not like P to F that happens all the time all over the world. So S to T that Phan Bik takes as diagnostic of Kuki Chin. So let's take just a word like tree we saw before is something like sin or sing in sign of Tibetan. So in Kuki Chin languages, you have tin or king. And that's a very peculiar change. So he sees that as kind of diagnostic of Kuki Chin. But the same change is also attested in other sign of Tibetan branches, Thongkul, Bodogaro and Bangru. And this is where like, I mean, no, this is why this is at the, you know, the forefront of research. It's not clear whether this is coincidence or it's contact induced or maybe all of the S to T languages are actually secretly part of the same branch. All of these languages are sort of between Northern Burma and Northeast India. Although in some cases with big gaps between them. So it's not clear what's going on there. And another case that is more clear but makes it a good methodological point is R changed to Y in standard Burmese and in Lachid. So this is the kind of a textbook case, right? You say, okay, R changed to Y in Burmese because it's affected Lachid and it's affected Burmese. But written Burmese still distinguishes R and it's translated as Y and the Erichanese dialect of Burmese still pronounces it as an R. So it can't be a shared innovation and it must be that probably it was a change that happened when Burmese already had a kind of prestige value across the whole region and that other languages kind of went along with the sound change through contact with Burmese. So this is a great example of it's, you notice that to closely read the languages share a change, you're tempted to posit that as a shared innovation indicative of the group but it's not always the case. And so it's always worth the testing whether or not it's the case. Tabeto-Burman as reconstructed by Benedict and Madesof had a simple five vowel system, I, E, O, U and old Chinese as reconstructed since the 1980s independently by three different research traditions has those five plus schwa. So then the question is, well, what's the relationship between the schwa in Chinese and things in Tabeto-Burman? And as I've handled proposed in 2008 that the change of schwa to a is characteristic of the Tabeto-Burman branch altogether. You know, that Sennitic has preserved the original sign of Tibetan vowel system that has, whereas Tabeto-Burman has collapsed two vowels together. And this gets to the kind of nomenclature issue that I was sort of avoiding earlier, which is it tends to be if you are in the Berkeley school, you'll use this term sign of Tibetan and then you'll say that that has two branches, Sennitic and Tabeto-Burman. Whereas the other scenes are more agnostic and say like, well, you know, there's no reason to group whatever 300 languages together as non Chinese. That is something that would take a lot of evidence to propose. But at least Zev, to his credit, tried to find an isogloss that would point in that direction. But it doesn't work. And this is something that I don't wanna take too much credit. I don't think I discovered it, but it's mentioned in my book, which is that in all Chinese corresponds to A in Burmese as in these examples. So we have fire, near and tail. Whereas I in all Chinese corresponds to I in all Burmese, like in sand, I don't know what overall definition to give here, but to be slanty or something like that. And something like, well, yeah. I mean, whether you're convinced by the semantics is another question, but at least it works with sand. So Chinese walk, lame, and then Burmese push aside or avoid shun. So this means that Burmese, in a very specific phonetic context, distinguishes inherited A from inheritua, which means that Zev Handel's proposal is wrong. And Guiyam Jacques has also shown that, oops, one more, that Tongut also preserves the distinction between Shah and A. So I think that's one sort of research theme is what is the evidence in non-Synetic languages for the A-Shua distinction, where in 2008, Handel said there won't be any because they all merged. But now it looks like, okay, they did merge in Tibetan, but they didn't merge in Burmese, they didn't merge in Tongut, and other languages it needs more study. So another one is uvillars. So several groups of scholars, so Piero-sen Starostin in Russia in the 90s, and then more recently, Baxter and Cigar, but particularly Cigar, have proposed that side of Tibetan had uvillars, and point to this correspondence pattern, where Tibetan has a velar, Burmese has a zero initial and Chinese has a uvillar, if you believe in those reconstructions of all Chinese that have uvillars. So I'm going to just set to one side that question, although I could talk about it if you want, whether or not all Chinese had uvillars, but if all Chinese had a uvillar, this is where it would fit in, and here's an example. So we have home in Tibetan, Kim, Im in Burmese, and something like come, subterranean room in all Chinese, is according to Cigar, the proposal, and then strangle or squeeze, suffocate, so kick in Tibetan ach, which comes from ick in Burmese, and then something like kicks in Chinese. And then here's an example to just show that the correspondence might work even when there's no Chinese edamon. So I think this is a solid pattern that it makes sense to reconstruct to the proto-language, whereas there's another pattern that I haven't given a slide for, which is all languages have k, all three languages have k, then you reconstruct k, and an example for that is the word for bitter, is like that, I think. So, but the Berkeley school, so Benedict already writing about the idea of Starosteen and Piedos really hates the idea of reconstructing u-vealers in old Chinese, sorry, in old Chinese and in Sino-Chinese, and Matasoff has also come out against it. And now on to morphology. There's two things I will discuss, causative formations and person agreement. So first, causative formations, in various Sino-Chinese languages, it's clear that there are two ways of forming the causative that are around. One is with an S prefix, where you have something like, I don't know, pat means to do something and spot means to cause someone to do it. And then another one is a litter, sorry, alternation between voiced and voiced in the initials, you'll have something like brad means to split and prad means to cut. And the traditional view, going back, I think all the way to August Conradie in the 19th century is to somehow associate these two processes, to say that the S prefix, is the same phenomenon as the voicing alternation, and the way that that is done is by explaining that the S prefix causes devoicing. It's pretty clear this does not work as an explanation. So Tibetan in particular has both kinds of causatives. So if the voicing alternation causative comes from the S prefix causative, then it won't work to have both. And then Giarong has two formations as well, the S prefix causative and a nasal prefix anti-causative, which is to say in like Japo Giarong, you would have something like prad means cut and un prad means to be cut. So Guillaume Jacques thinks that it's that pre-nasalization that causes the voicing. So it's exactly the opposite of the, let's say the traditional explanation is the S causes the devoicing, and Guillaume Jacques's explanation is that the pre-nasalization associated with the anti-causative causes the voicing. So just to give an example, we have this Burmese qia to fall or to drop versus Japo Giarong un gra. And then we have Burmese qia with an aspirate qia to bring down or lower and Japo Giarong kra. So you see here that the association of the pre-nasalized form in Giarong with the non-aspirate form, which comes from a voice form in Burmese. So that's how, let's say the, this is an example of a kind of disagreement between if you like the Berkeley school and the Paris school here. And then turning to agreement, the better studied languages don't have agreement, right? So Tibetan doesn't have agreement, Burmese doesn't have agreement, Chinese doesn't have agreement. And that actually led to this sense that Antoine Maillet writes about that doing historical linguistics on East Asian languages will be hard. There just isn't the kind of machinery in those languages that gives you things to work with like morphological alternations like agreements and interesting compounding, whatnot that we're used to in the European. So Maillet, the spared that reconstructing sign of Tibetan would be possible, but as more languages with agreement have been described, we have two possible explanations. So, or I should pause there and say, now sign of Tibetan is quite interesting in terms of having some very isolating languages with simple morphology and some extremely complicated languages. So Kiranthi, Gyarong have beautiful agreement systems, complicated constant clusters, and then, Karen or Lolo Burmese have very simple sort of typical Southeast Asian sort of agglutinative morphology, lots of light verb constructions, simple syllable structure. And I think generally speaking, a language family either has one typological profile or another that's associated with the typological profile of its ancestral language, but sign of Tibetan, we have this kind of split. So there's two explanations that are possible. One is that languages with agreement share an innovation and this is what Lopola thinks and he calls this family the sub-branch or rune which where he associates all the sub-branches together that have agreement. So he would branch together Kiranthi and Changiq, for example, into rune. And then the other view out there is that languages without agreement have more or less independently lost it. And that's a view that Scott Delancey has been a very vocal advocate of. Back in the 90s, actually I think it was mostly Lopola versus Fandrine, but at some point, I don't know, Fandrine decided to spend his time in other ways and it became sort of Lopola versus Delancey. And depending on what your appetite is for academic controversy, you really can read 20 different articles where Lopola says this and Delancey says this and Lopola says this and Delancey says this. It's all rather undignified. And personally, I think there's been more heat than light created by this controversy, but over time, I think the descriptive work is favoring Delancey's position. So, and then I just want to mention that I think Kuki Chin stem alternation and Yalronic stem alternation is a promising area of future research. The issue, part of the issue is that Lopola sees that pronouns are similar to affixes. Let's say we're talking about Greek and we have a verb like hisse-meh. Lopola would say, well, meh is very similar to the first person pronoun that in some of the oblique stems is something like meh. So probably the Greeks just stuck their pronoun at the end of their verb, yeah. And I think that actually, you know, let's say that's what Vap said about Indo-European. And I think that to some extent that's how languages get agreement system. The question is at what time depth you're talking about. And it is clear that to a certain extent parts of different agreement systems in scientific event are recent and have been supplied by pronouns, maybe to some extent, analogically, which is why I think stem alternation is a better place to look to kind of thread the needle of this controversy because you can create suffixes analogically based on pronouns, but you don't do that with stem alternation. So if you can show that a language has an agreement system in terms of, oh, we use stem one in these contexts and stem two in these contexts that have to do with agreement, then that's a way of avoiding Lopola's criticism that it's all grammaticalization pathways. Coming to syntax, the most important point to make is that it's not really clear that you can do comparative syntax in terms of reconstruction. So even in, so, well, first of all, to do comparative syntax, you would have to have pretty good comparative phonology and comparative morphology already worked out. And we don't have that for some of them. But even in Indo-European, the methodology and if you like the possibility of comparative syntax remains insecure and controversial. And the issue here is that it's very hard to distinguish borrowing and shared inheritance when it comes to syntax. Whereas for phonology and morphology, we have these sound laws, these kind of systematic correspondences that help us separate out what's new and what's old. But there doesn't seem to be anything like that for syntax. So it's very hard to do a syntactic reconstruction. And as I said very early on, there isn't much of a tradition of philology and publishing of texts in sign of Tibetan languages. So oftentimes, and actually I could even use the example of Kugi Chin here, you say, okay, I found a language that has really interesting stem alternation that has to do with whether something's subordinate or finite. So I'm going to compare across the family, which languages use stem one, stem two. And then you realize that for a particular language, you only have two sentences in a text, in someone's PhD, where they say, oh, you use stem one like this, you use stem two like this, and it gives two sentences each. It's not enough to draw any conclusions from. And the fact that text publication has been so under-emphasized in sign of Tibetan linguistics means that there just isn't very much to work with. And for syntax, you need texts, yeah? So I think comparative syntax is basically impossible and won't become possible in my lifetime, but I still want to say that there are some obvious questions out there, like why are Chinese and Korean not verb final? Every single sign of Tibetan language is verb final, except for synodic and karenic. Karenic is usually explained that that's because they're so far south and they must have been in contact with other people who put verb in second position. That explanation doesn't work for Chinese. So how did, presumably Chinese was verb final, switched to putting verb in second position? Personally, I see the use of yee as a kind of copula at the end of the sentence as indicative of that, but that is controversial among sinologists. And it is worth saying that Chinese has been verb second since the very beginning. Oracle bone inscriptions are already verb second. So anyhow, that's one question, is how did Chinese and Korean end up so different? And then another question I think that's kind of typical of this part of Asia is how to handle clause chaining. And I would say that part of the kind of functionalist, typological orientation of most sign of Tibetan field workers is they love to say that things are the result of grammaticalization and treat that as if that's done. You say, oh, I don't know. Let me think of, you know, oh yeah. So the future in French, you know, je chanterai, oh, that comes from Kantare-Habeo and is a result of grammaticalization. Well, that might be true, but it still doesn't tell you, you know, how to use the future in French, right? And there tends to be, I think, a practice among descriptive linguists of sign of Tibetan languages to kind of mix up synchrony and dichrony particularly when it comes to grammaticalization. So they sort of say, oh, look, this is probably grammaticalization and treat that as if it's a synchronic description. And that's something that annoys me. And then I'll just say that my hunch is that there's a lot of grammatical stuff going on in clause chaining and proverbial morphology that has yet to be discovered and to just sort of point towards the fact that I might be right, I will point out that Tibetan, classical Tibetan, has a robust switch reference system, which is, you know, in case you're not familiar with switch reference systems, this is you have clause chaining where you would say sort of like, in English, might say like, having gone to the store and seeking to purchase milk while, you know, this like long series of subordinations using non-finite verbs, but the choice of verb ending indicates whether or not you're talking about the same person or you're switching to talking about someone else. And in Tibetan, it's very clear that like you use this one suffix if the story continues with the same subject and you use this and not other suffix if you're switching to another person. And that system was let's say sort of discovered around 1988, but the publication that described it was described in extremely confusing terms and was only sort of clearly written about in 2019 by Zach Beer and still hasn't hit the textbooks or textbooks and grammars. So in Tibetan, which has been worked on for 150 years, there's a pattern of, you know, there's a key grammatical feature of the language that was only clearly described last year. So I think that across sign of Tibetan languages, clause chaining and the sort of syntax of clause chaining is an area that is just ripe for more research that potentially would be very insightful. Looking to the future, and this will presage some of my lectures from tomorrow, we don't have enough people working in sign of Tibetan linguistics. That's, I think, the problem. You know, there's, think about how many people work on German or Italian or Greek or Latin or even Chinese or Japanese. Well, there are 300 languages in China. And, you know, opera and moa, each one should have a department that studies the culture and language and literature of that people, but it's not gonna happen. And we just have to accept that. So for sign of Tibetan linguistics to achieve things comparable in impressiveness to what we've seen happen in European linguistics, we just have to become more productive. And the only way to do that is through technological innovation. So that's one of my kind of ide fix, if you like, is that we need to embrace technological innovation, whereas people who work on Greek and Latin, maybe they don't, because there's so many of them, they don't need to speed up their work. But we, in sign of Tibetan, we do. And I think that there are some technologies that are available for doing that, that I will describe in more detail in a future talk. So automatic transcription of sound files into IPA, automatic glossing translation and aid selects a call for using off the shelf natural language processing tools, and then some new techniques in automating elements of comparative linguistics that we'll just, for the moment, call automatic cognitive detection. And I think that each of these, so I see this as a sort of pipeline, and I'll return to this tomorrow, but you want to go from having sound files of some toothless granny telling you a story in a hill somewhere to a dictionary and a grammar and research papers. And we can imagine a kind of workflow pipeline that changes sound files into dictionaries or something like that. And I think that every major step along the way, there is a technology now available that would speed that up. But we're still a long way off from kind of engineering that pipeline. Yeah, so we have all the sort of pieces, but we need to kind of glue them together correctly. And I think the most important thing is to start incorporating a greater technological sophistication into the training of linguistics students. And I think this is happening already, but could be happening in a more self-conscious and productive way. And that's something I'll also talk about in the future. And then let's say the paradox from my side is, I can definitely see that this is what needs to happen, but I'm something of a Luddite, at least at the practical level myself. So I feel like I can kind of tell you what I see on the horizon, but it's the 30 people who are attending this today who will be more likely to actually implement that than I am.