 What I have in mind is to briefly survey the progress that has been made since the mid-90s, which was when I first began working in Kukicin languages, in our understanding of Kukicin or South Central Tibetan Berman for the domains of alignment typology, and then in so far as alignment typology doesn't impinge on the verbal complex, some other aspects of the verbal complex, so valence affecting phenomena, directionals, and what I've called verbal classifiers. And if I have enough time, I think I will, I also mentioned elaborate expressions, which is something we may already be too late for, unfortunately. So Kukicin or South Central includes the languages in one, and I won't belabor that since I think probably everybody here is familiar with the subgrouping that's given there. So I just include the languages so you can get some idea of the different parts of the group that I'm talking about. So moving straight to alignment typology. By the mid-90s there was, I believe, a widespread understanding of South Central languages as being urgative, at least with regards to case-marking. And at that point, a few verbal participant-marking systems had been fully analyzed, so there's less certainty about how those systems tend to pattern. So two illustrates, the example two at the bottom of page one illustrates core case-marking in MISO. So 2A shows an intransitive-subject goat, Kale, without any case-marking. And if you compare at the top of page two and hand out 2B, you'll see that a goat in that sentence or the goat in that sentence is marked by the suffix in. So it's a transitive subject there. The transitive object into B, grass, like the intransitive participant into A, is unmarked for case. So an urgative-absolutive pattern, essentially. And 3 and 4 show similar facts for Lai and Tedim, two other languages that we knew quite a bit about by that time. And so that was our impression of what Kukichen languages were like in terms of case-marking. In terms of verbal participant-marking, we had been exposed to systems like the one in Lai, illustrated in 5, where the verbal participant-marking is not urgative-absolutive in terms of its alignment. So Lai shows essentially nominative-accusative-marking in second and third persons, but neutral-marking in first person. So if you compare the first singular forms for intransitive-intransitive subjects, so S and A, versus P, the object in transitive clauses, you have no distinction between the first person singular markers, at least in terms of form. They're placed slightly differently in the order of elements, which distinguishes them. But formally they're the same, at first person the same thing. But you'll see there's a difference between A and S in the case of the second and third person markers and P. So essentially nominative-accusative-type alignment there. MISO down at the very bottom, which we don't need to look at in detail, it's a little bit small on the handout, had a different system, but still not too far from straightforward nominative-accusative in certain respects. Certainly no hint of ergitivity or anything more unexpected. So Bedel, who amassed a body of treatments of several systems over the last couple of decades, reached, well, beginning in the 90s, reached I think his high point with Kho, which I extracted the forms from an article of his in number seven at one point. And if we won't examine them carefully, but if you look through them, you'll see that there's no indication that there is ergitivity in the marking there. It's essentially nominative-accusative, although there are some some some syncretisms and other complications to the paradigm, but nothing out of the ordinary. But in Delancey's discussion of Mara in the paper that touched off years of controversy in Tibetan-Berman linguistics regarding the reconstructability of pronominal marking, still ongoing, he hinted at hierarchical marking in Mara, which nevertheless was at that point undergoing a restructuring, so not completely recognizable as a hierarchical marking, particularly an inverse marking system. But that was a hint of things to come. So in the present, we, 25 years later, certainly have seen a large number of other ergative-absolute of case marking systems described. So Dai, Hyou, Helga Hartman's account of Lemmy treats it as being essentially ergative. Maybe there are some which have less clearly ergative-absolute of systems. So a lot of the older sources for these languages don't indicate strict ergative-absolute of case marking. But I actually, Hyou, I should mention, it doesn't have strict ergative-absolute. It's like you see in Mizo and Lai, it's got split ergative marking with third person marked, and second and first person unmarked in terms of over nominal elements. So other systems, maybe there's something that we could describe it as a gentive marker, but maybe it's not strictly speaking an ergative marker. So maybe Hyou, but I can comment about whether Lemkanya's like that or not, and I'm not sure about Monsang. But it would make sense that a lot of languages of Manipur might end up with more agent marking rather than ergative marking due to contact with Matei. So that's one side of things where things have gotten a little bit less straightforward. At the same time, we've also got a number of examples of languages which have nominative accusative, essentially case marking boringly in, for instance, Kumi. The elements that are underlined there are, those are tonally marked as locative. So the locative marker is extended to P participants in Kumi if they're sufficiently salient as they are in Senate 8. In 9, Rangmicha, like Kumi, also has extension of locative marking to P participants if they're sufficiently salient. And there's a segmental marking there. So this element you see in 9, in 9 it actually marks not only the matnit element, which is the P participant, the people who were floated off on the raft. Also that example which shows you just a straightforward instance of the nomarker as a locative marker. The no at the end of the sentence you'd be tempted to call that a locative marker as well. But no actually is in this case a borrowing from Roo. Rangmicha marks sequentiality with other means, not them the element. So in terms of participant marking at the top of page 4, a number of systems have been recovered, which at least in part involve hierarchical specifically inverse marking. So in 10 from the treatment in the excellent volume from Himalayan linguistics on northeast Indian person indexation systems, you see that there is an M prefix which occurs in the lower left-hand quadrant of the paradigm, where you have second persons and third persons acting on first persons and third persons acting on second persons and first persons. So just in the appropriate conditions to act as an inverse marker, reversing the default interpretation of person markers mapped to roles according to the hierarchy 1 over 2 over 3, which is apparently operative in the language. In Hyo, Zakaria has shown that you have a knee inverse marker which makes in a number 11 a number of appearances because in this language first person is not ranked over second person, but first person and second person are ranked equally in the hierarchy. So whenever you get the knee marker whenever second person is acting on first person, but you also get it when first person is acting on second person. And more confusingly, knee also, if I'm not mistaken, indicates plural in certain forms in this paradigm. So another knee, which is often with the inverse marker, also pops up in a number of places. More straightforwardly, in Lamkang, at least for the portion of the paradigm that's seen in number 12, T marks inverse under conditions which are comparable to those which were seen from Lonsang. So again, it's in the lower left-hand quadrant. So this T prefix marking an inverse situation. So two related issues where considerable progress has been made in the last quarter decade include recognition of the prevalence of post-verbal participant marking paradigm remnants, if not, you know, pretty full paradigms. So particularly in the Northwestern languages, formerly known as old kooky languages, and also in Southeastern languages like Hyo, not illustrated in number 11, but Hyo has in the negative pretty robust representation of the what must be reconstructable, post-verbal participant marking as Delancey has discussed. And there's also the prevalence of systems where the verbal complex is more fragmented in general with interspersed auxiliary or auxiliary-like elements rather than forming a coherent agglutinative piece. So I mean, it may still form an agglutinative piece, but there are portions of it which are analyzable as in containing auxiliary elements. So that is indeed progress, I think, in terms of our understanding of the alignment of these languages. Turning to three, some other aspects of the verbal complex. From the discussions that were available for LISO and BOM and MISO, TEDM, CISANG, by the mid-90s, there were at least fragmentary discussions in older sources that were indicative of what might be there. By the mid-90s, we realized there were morphological causatives and possibly benefactives for some of these languages. There may be as evidence in the discussion of BOM by Reichler, possibly also for LISO of other types of applicative-like constructions. But what we know after 25 years is that there is a widespread suffix or causative sock, recognized long ago, but only four specific portions in the family. But it seems to be pretty widespread. So it's found in Dai, in Shaq. It's found in the productive causative shock in Hyo. It's marginally attested in Ring Micha and less grammaticalized than I would expect it to be in Ring Micha, given its presence elsewhere in the family. Even in Lai, it makes an appearance in a few lexicalized verbs. In LISO, it actually turns up as a benefactive marker rather than a causative, but that also makes sense. Probably older is a P or M prefix or causative, also widely attested, found in Dai, Kumi, Ring Micha, Sorbonne, Lamkang. I forgot to check, but I think it's also found in Pankwa. So 13 gives an example from Ring Micha of this prefix. And so it gives the prefixed causative and blot, and then it also gives an instance of that verb without the prefix in an intransitive sense. So that's that one. At the top of page six, there's also, of course, initial voiceless stop aspiration in Sonoran, devoicing that it's widely attributed to an S causative prefix seen elsewhere in Tibet or Burman. This also has a fairly broad distribution. So in Central Chin and also in Southeastern and Hio, it's very robust in Hio as Shrumbo Zakaria. And finally, there are various more sporadic developments like the causative der in Lai, which I think is also tir in Mizo, and then there's a Southwestern element, Hai, which I'll talk about in a second. So regarding applicatives in 1998, I wrote about the remarkable family of applicatives that's seen in Hakka Lai, of which I'll draw special attention to the benefactive null-affective applicative in 14A, which like as far as I know, almost all benefactive null-affective applicatives comes from the verb to give. The committative in 14C marked by B, the relinquitive in 14F, dak, and the instrumental in 14G marked by Nak. And I draw attention to these because these are, I think, the most widely attested. They have the widest distribution in South Central. So for instance, in 15, at the top of page 7, all of these show up in Dai. So there's a benefactive null-affective marker based on the verb to give. There is a relinquitive element identical to Lai's. The committative has bui, which must be the predecessor to Lai's bi. It's a regular development of the ui, rhyme, and lai. And the instrumental marked by Nak Hyo exhibits most of these as well, seen in 16. So three of the four at least. And ring-micha and kumi, on the other hand, show somewhat more impoverished system with only a bui, a benefactive null-affective applicative attested in ring-micha, as shown in 17. I neglected, I'm sorry, to include an example of the kumi marker, but it's marked with B, virtually identical to ring-micha's in its behavior. And there is also in these two languages an element hy, which I can talk about the grammaticalization source for that later if you ask me about it. But it marks instrumentals and committatives in ring-micha, as in example 18, but in kumi it has a greater functionality. It marks kumi's productive causative construction, or it can also mark various types of applicative construction, including instrumental applicatives and goal applicatives, exemplified in 19 for the causative, and 20 and 21 for the applicative constructions. So another valence affecting construction type, which we knew would be important probably by the late 90s, was the middle, but we're only now becoming more aware of constructions of this sort. And I think we'll hear more about them, at least in one language this afternoon. So maybe we'll hear about them. I'm not sure what we'll hear about in Tsumtu, whether it's a middle there or not. But the middles are important. And for one thing, because there is this outlier middle marker that occurs in southeastern languages, but also they're interesting because of their potential relation to things which are involved in the inverse systems. So potential relationships of inverse marking and other middle-like phenomena draws the inverse marking that we see into possible light in the context of middles. A further aspect of the verbal complex worth considering is directional elements with the top of page 8. So by the mid-90s, attested descriptions of directional marking involved predominantly prefixal systems. So the descriptions we've already discussed for central languages like mezzo, bomb, and lizo. And the LSI had little tidbits of information from lots of languages, most of it involving prefixal elements. Hartman's 1989 treatment of die directionals was the exception. There were hints that there might be more elements occurring after the verb as well. So skipping over 22 for a second and going to directionals now, we have an update on So Hartman 1989 in the form of Helga So Hartman's dissertation from SOAS on die, which turned into her grammar of die, where she notes important distinctions in directionals depending on the motion or lack thereof of the agent involved. So that's brought in as a parameter besides some of the other parameters which are detected widespread. So in 2014, I surveyed available materials and made the observations on directionals in 22 that the prefixal elements listed there are or widespread. But then there are also some preverbal or prefixal elements found. Sorry, there are also some suffixal or postverbal elements found in southeastern and southwestern languages. And I said that maybe there are some exclusively postverbal suffixal languages, but I don't think that's the case because all languages that have suffixal ones have something preverbal at least. So they both occur. So we since the 90s also have papers by Chellion Ut on Lam Gan, paper by Van Beek and Klangnet, and we're fully treating the elements in lie. So these are preverbal systems which have a lot of similarities to other systems that have been described. Zechariah's description of Hyo's directionals include two elements, unless I've missed one. So the preverbal element which resembles an andative and a postverbal one, which also resembles kind of an andative, an all element that Zechariah calls a departative. So not a very extensive system compared to other languages like Dai in the southeast, in the southwest. The systems are not generally that rich, but Rangmicha appears to have a fairly rich system in part because it borrows some elements from Mru. So alongside elements which are native Kugichen elements, it uses an element Jam, which is an andative. It uses Dkut or Kut, which is an element that means back or again, which also is apparently from Mru. But it also has elements of its own that I've worked on recently, including a number of things that are andative that I list in 23. And the examples I give on the handout don't illustrate all of these. So Hu actually, it turns out, is the most neutral of these andatives. I'll talk about Kui in a second, and I haven't got an example of Bai here, but I actually understand how Bai is different from the others. So I can tell you about that if you would like me to later. So Rangmicha and Kumi both have venatives, which presumably reflect the archaic venative, which I said probably, well, I'm not the only person who said it. So Delancey probably said it has a form something similar to Hval Ng or ng. So usually it's only an ng prefix in Rangmicha as in 24. In Kumi it has a larger form ng as in 25. Rangmicha also has what I've determined to be at the top of page 9, a distributive, or maybe I should call it a distributive andative, distributive andative, sounds strange. So a distributive andative is what I've settled on. So it indicates motion away from a dyctic center, plus action performed in various places, or in order to perform action in various places, like you see in example 26. So this Kui elemented Rangmicha is actually cognate with a perfective marker, Vui in Kumi, and I'm still assessing the extent to which it retains any residual distributive semantics in Kumi. I haven't detected it before, so I'm trying to recheck it, but I don't think that it does. Rangmicha and Kumi also have apparently unrelated upwards motion directionals which are post-verbal, seen in 27 and 28. So kang in Rangmicha, and galaw in Kumi. Also post-verbal are Rangmicha and Kumi's downwards directionals, which actually are cognate seen in 29 and 30. So probably you can tell that tuk in Rangmicha is feasibly cognate with katiu. So there must have been a prefix on this element that Rangmicha's lost, but the iu rai in Kumi reflects an uke rai in Rangmicha, or in their predecessor. So I'll also note Conrith's work on the relationship between venitives and cislocatives, or venitives or cislocatives as she refers to them, and participant marking, which I believe has a wider distribution than just in northeast where she notes that it occurs. So development of second person marking out of a venitive or cislocative marker. So similar sorts of things I think are going on in Kumi with their venitive, maybe also in Rangmicha, although it's not clear. So there's still more work to be done in terms of directionals, but we're obtaining, I believe, a critical mass of information in this area as well, just like I think we have for valence effecting instructions. So it's turned to a couple of areas which have maybe not had as much progress in them. So what I refer to as verbal classifiers in South Central. So Henderson had this notion of chiming. So these were adverbial elements which occurred in postverbal position and had a form, I guess, approximating something like ding-dong or cling-clang, hence the name chiming. I'm otherwise not really sure where the word chiming came from, unless it was a translation from tedium or a translation from Burmese, I don't know. There's a sort of rhyme in chiming. It comes from the descriptions of Burmese adverbs. Uh-huh. Oh, okay. Yeah, so anyway, Bhaskar Rao had told us more about tedium chiming by the mid-90s and patent relabeled these chimes. Idiophones, as demonstrated by my favorite example of these in number 31. So in 31a, the element watma occurs and evokes a the image of a large or rotund baby, I guess fat baby. Yatma, on the other hand, creates the image of a baby which is small or thin. And there's a typo in that, so it shouldn't be nanakcha. It should be nakcha with an h rather than an n in 31b. There's also typo in the gloss for elephant in 32, which I'll turn to next. So 32 and 33 show you the related phenomenon in kumi where the elements in question focus mostly on the size of the referent that's involved. So there are somatic nuances that often co-occur with these elements. And they may involve some element of visual imagery or other kind of imagery like they do in tedium and lye. But the sense of largeness or smallness is really the central element of meaning in the mit would seem. So for instance, in 32 the element ga refers to a relatively large king's daughter as opposed to the element ge in 33, which refers to the smallness of the child that's involved here. And in both of these instances, the ga or ge element also includes the information that whatever verbing occurred, it verbed the relevant referent to death. So the king's daughter dies in 32 due to being stepped on by the elephant. And in 33 it's this spirit that I refer to as an ogre in other work which captures his son and beats him to death and then eats him and plants his head atop the the granary spoke or stick or whatever I guess it is. So anyway, Reg Mitchell also has a verbal classifier, which is actually identical to this kaka one in kumi. So again, in 34 comments on the large size of the sister and that she is, well, bop is the neutral word to mean to kill. So she's killed to death, I guess. Superfluous, but the element of to death is still there in this one as well. But not all of these elements have lots of rich additional semantics other than size. So hoop loop and hoop loop can refer to both the motion of a dove or an elephant. But hoop loop is a small or diminutive verbal classifier in reference to a dove in 35 a but hoop loop is in reference to an elephant in 35 b so a large or an augmentative verbal classifier there. So those occur also in Reg Mitchell and where they're tested, as I've already said, size seems to be the most central sense that they have either literally or metaphorically. So it can sometimes refer to relative extent of some quality if the predicate refers to a quality rather than a more active event. So they may appear in non-reduplicated form in kumi and during Mitchell, at least I'm pretty sure also in Mu. So they don't really have a chime formed to them. They also don't have a change in the vowel in almost all cases. There may be a couple of exceptions to that in kumi that I can think of. And again, they often do, although their central notion is size, bear some idiosyncratic somatic nuances or can be associated with particular imagery resembling idiophones. And this letter observation is important for our development of lexical resources because where these occur it's really important that you actually check for every verb what which of these can occur with the verb and what effect do they have. And I haven't, I must have done that yet for either of these languages and it's a daunting task that hopefully native speakers will help us with. However, since I made these observations about chiming and verbal classifiers a decade ago, not much has been shown about their use elsewhere in kuki-chan. And I'm worried that their use may be waning if it wasn't already waning by the time I made the observations that I made about kumi. And I know I'm running probably pretty short on time, but just briefly on elaborate expressions, I think this also holds for elaborate expressions illustrated in 3B with the very first elaborate expression in kumi I encountered in the wild where a tiger is searching for a girl who's hiding from him. So the 2D2D and the Leiloh structures there involve an elaboration element. So pudung and long are enough to mean water gourd or pot. And they've got this default reduplicative templatic elaboration of them. Not all elaborations are reduplicative, that's just the kind of default. And this has very similar sorts of effects to what you see for elaborate expressions elsewhere in Southeast Asia in terms of making the discourse or the utterance sound nice, and also in terms of some other things that I've suggested correlate with the use of these. Ringmage also has elaborate expressions. So for instance, bangdong in number 37 appears to be an elaboration for shoulder. So I can't for the life of me get consultants to tell me that it means anything. And after long discussions have decided it must be what I'm calling elaborations. So there are also aspects to this example which make it fit in with other kinds of elaboration that I can tell you about if you ask me about it. So the ringmage and other languages in the area like Muru also exhibit elaborations, elaborate expressions, not as robustly as Kumi does, but nevertheless relatively frequently. So, and it's, as I've alluded to, unclear how widespread or robust this phenomenon is. So little description of it has been forthcoming since I worked on this about 10 years ago. And nevertheless, again, the unpredictability of elaborate expression formation is potentially significant for the development of lexical resources because ideally for anything that can have an elaborate expression you would have a recording of what the elaborate expression was at least. So let me just conclude briefly by observing the obvious, the progress that's made in the analysis of Kukishin Morphosyntax over the last 10 years. It was everything to people getting out and expanding the corpus of data for Kukishin languages. So something that we are all participating in and can be proud of. But there's clearly much more to do and I hope to have demonstrated some areas where I think we're doing a pretty good job and pointed to a few other places where we might dig a bit deeper. So thank you. Thank you very much, David.