 And I'm particularly happy to introduce our first speaker today because of the name is Tanya Slavin and she's a new research associate of departments starting this year, so she'll be here for us for a while, at least. And she also works on very interesting languages, something we do not really study very much here at SOAS, namely Algonquian languages spoken in Canada, she's going to talk about it. She did her PhD in Canada at the University of Toronto, working on one of these languages, called OJ Cree, that sounds welcome to hear today, and worked on morphology, syntax, semantics, in particular on the structure of steps. And I understand that it's the topic of her PhD and that it's also the topic of today's talk. So please welcome Tanya Slavin from SOAS. So as I've already said, most of my work for, yeah, almost all of my work, it's OJ Cree. It's a dialect of Ojibwa, Ojibwa is an Albanian language spoken in Canada, and OJ Cree in particular is spoken in geographic-resolated areas in the province of SOA. Is it loud enough? No. So OJ Cree is spoken in geographic-resolated areas in northern Ontario, province of Ontario and province of Manitoba in Canada. So Ojibwa is, like many Albanian languages, has a very endangered language. It's a dinar, many, most of its dialects are not spoken any more by the large amounts of people, it's just by handful of people. OJ Cree in this sense is a little better off, possibly because it's geographic-resolated. You can only get to most of its communities only accessible by planes or by winter over the winters. It's fun because of that, the dialect is quite well preserved. It's also, for the same reasons, the least described dialect of Ojibwa, there's no grammar or OJ Cree, it's not very well documented at all. But it's also, for me it's one of the most interesting dialects, because it retains some of the productive dialects that are lost in other dialects. These are things that I'm interested in, the policy of the system, and which are more logical, the patterns. And this is what this talk is about, this talk is about the structure of the verb stem of this language. So basically the goals of my work is to gather large amounts of data to aid the documentation of the language while analyzing these structures. So as I said, it's a policy of the language and the verb stem is usually composed here of more than one verb stem. So in one A you see the template for the verbal complex of this language. So the stem, the part that we are going to be concerned with in this talk is involved, but there's various other components of the verbal complex. So we start with the personal prefix, followed by a tense marker, and then one more adverbal modifiers, pre-verbal modifiers, pre-verb to the code, and the kind of adverb of elements, then the stem, and then the rest of the ingredient. Here's an example of the verbal complex that has all of these elements. Niki, kimuchi, papa, meta, panani, totem, and it's a secret job of my friend around, and it starts with the personal prefix, li, followed by the past tense marker, ki, then the adverb, kimuchi, secretly, then you can have as many adverbs as there's one here. Secretly, yesterday, for some reason, yeah. Then the stem, papa, meta, panani, and an agreement in production, indicating that this is first person, acting third person. Now, so we're interested in the structure of the stem. The stem itself, papa, meta, panani, here consists of more than one morphic. In this case it's a final philic. It may contain more than a three or more morphics. So the question is, what is the structure of the stem itself, and what is the relationship between, how do we know that the stem, because it's the, so papa, the element papa, it means drive by itself, but it can't stand on its own. It can't start in inflection. You have to say, I'm driving here, he's driving, you have to have the papami part, so these two parts work together, papami, then. Now, so this question has concerned Algonquinist for a long, long time. What is the structure of the stem? And Chris Wilford, in one of his papers on Plain Scree, in a little language, summarized this question very well. Here's what he said. Since the verb stem in Algonquin languages usually consists of more than one morphic, two questions have long been of interest to Algonquinist. How should this morphic be classified, and what are the relations among the resulting classes? Kroger in 1960 wrote that the undeterminate fundamental problem of Algonquinist is whether his language is saying he enters looks, or he enters lookingly, or enterlingly he looks, while the problem has been under consideration, at least in the time of people. So it is not much closer to satisfactory solution today than it was in Kroger's days. So that's the question. How do we analyze the structure of stem, the relation between its components? Now the traditional view, the traditional description of the stem structure in Algonquin languages was written by Bloomfield, and he described the stem as consisting of three elements, basically according to a pre-defined template. The sediments are called according to their position relative to each other. So they're called initial, the element at the beginning of the stem, middle is whatever is in the middle, and final is whatever is in the end. So here are some examples of the stem, of stems and how they would fit into this template. So Tafki siten, she has cold feet, has all three elements. So at the beginning of some kind of vegetable element, Tafki, then middle is usually some kind of incorporated nominal, and final is a category-defining element. So here it says that it's a transduper text, and it's such. Now the next one, Tafki, Lafitte, someone has two of these elements, the root Lafitte and the category-fining final. So Pa is the initial in this template, in this diplomatic approach, and V is the final. Another example is Ompi-ki. She grew up Ompi and Ki, the initial final, and then you can also have stem, the contest upon the Lafitte. One element, V, for instance. So this is a U that's accepted much universally by all subsequent scholarship. So the stem, stems in Abadmala language consist of a built-in template in the initial media plan. Now that's a useful descriptive device, but it doesn't tell us a lot about the structure of the stem, what's going on inside, and what the relationship between its components is. So in my work, and it's been also done by many, by other people, implicitly or explicitly, they try to move away from the template and figure out what's going on. So in this talk, I'm going to look at two different types of stems, so two unrelated constructions. In the first part, I'm going to look at Lafitte's five-part stem, and the second part, I'm going to look at stems that involve non-properation, possessing non-properation. So it's two different constructions. Basically, they're going to have different analysis, but they both show you that there's no way you can feed a stem into a certified template. You have to analyze them in different ways. So we're pretty dense in terms of data and analysis, but hopefully we'll give it the same. So bi-partite stems. In these stems, call them bi-partite, because the root in these stems has to be preceded by something else, some additional verbal materials. For instance, imparto means run to itself, and you can argue whether it's contested parto or just parto. There's no agreement with that. But parto means run to itself, but you can't just say it's running parto. You have to add image, run along. The same as the case in 3D. Peach human carries something over. Women carry, but you have to say have peachy. You carry over here. And minoshimo, you can just say shimo. You have to say minoshimo. So the question is, why is this verbal there? Why do we have to have it there? To have it for the stem. So we can say that bi-partite stems a subject to a left-age environment. These environments have been used in all kinds to change life a different way. But I'm using it here to describe this pattern. So this is the question that's going to concern us in this part of the talk. What is the verbal element? Now I need to remind you of the position of the stem in the verbal context. So remember that the stem is preceded, optionally preceded by a verbal, it's called pre-verbal modifier, or I'm going to sometimes call it pre-verb, or stem-external modifier. Now, similar to the constructions through the five variable, the pre-verbal modifier, classical stem. Here the pre-verbal modifier is the same element as the left-age element in bi-partite stem. So when I say left-age element, I mean this is internal or normal, it has to be inside the stem. So if you compare these two minimal pairs, they're almost non-renewable. But pimi-parto, in the case of pimi-parto 3D, pimi is a verbal alarm, it's inside the stem. But pimi-nicamo, in 5A, inside the stem, if you put the word say nicamo, it's the same thing. Pimi is optional here. In 3B, pichi is inside the stem, pichi when carried over here. You can also have pichi as an optional verbal, pichi papamosi, 5B, if you're working with the direction, you can just say papamosi, it's fine. Yeah, the same in all of this. Now, this is what I call left-age apartment. So in the extension 3, the verbal has to be there. Now, another way to look at this, about this question is, what is the difference between the constructions of 3 and the 5? Why is it that 3 is obligatory when 5, you don't have to have to care? There's been various proposals about this requirement. Saying that it's an arbitrary morphological property in the word prosodic requirement or syntactic requirement. So I don't have time to go over this. But I will explore that because I know that the left-age requirement is actually a semantic requirement. It has to do with the left-edge. So basically, the element in the left-edge fills the gap in the semantic loop and completes the left-edge. I forgot to say that I call these roots weak roots to distinguish them from a different type that I'm not talking about here. So this looks more like a better term. Okay, so I'm going to first, to a general overview of what can appear on the left-edge, basically to give you an idea of what can appear here. And then I'm going to present a proposal and test this proposal with various predictions. Okay, so here's one example of a weak root that can be perceived by various left-age elements. So mo, means peak, x, speaks some language. But you can't use mo by itself. You have to have something there. So nishnade mo, she speaks Uchukvi. Or you can say Ujibwe mo, she speaks Ujibwe. You can say aganashi mo, she speaks French. She speaks English, or Ujibwe Tikoshi is a noun, it's a white person. So these are all nouns that appear on the left-edge. Mo, she speaks one of the languages of overseas. Nishnade mo, she speaks a foreign language. It's a very, very productive pattern. It's not an exhaustive list, for example, of any way they will appear on this very, very productive pattern. So we can say that mo, means something like speak a language off in the left-age element, fill the gap by supplying the name of the root. Another root of the symbol of meaning is used language in a certain way. Kishiwe. So here we can have a wider range of elements. You can say wakusin kishi kishiwe, she speaks the right language. Where ishi is a kind of linker that connects the Ujibwe to it. You can say Nishnade mo, she speaks Ujikui. Now here notice that in 6A we had a Nishnade mo and the a Nishnade is a noun means Ujikui person. But a Nishnade mo, it's the Ujikui language. So you have to have a noun for language here. And this one is also very productive. Akamihisigami kishiwe, she speaks a language from across the sea. As in 7C. But not only for languages. You can say as in 7G in the next page. You can say machi kishiwe that would mean to use bad language for minokishiwe. Kishiwe means use language in a certain way. And then the left-age element here fills the gap by providing language a quantifier or a verbal of manner to pretend. So hanokii krasso, this one takes any kind of verb or any kind of noun. So I can say she tends to work hanokii krasso, she tends to sleep ni bak krasso, I'm given over. You can look at the examples later. You can have a complex verb. When I wish ni bak krasso is 90. When I wish ni bak, she has no place to sleep. You can have a verb instead of its own reverb or its own bars. One very interesting example that I'm hoping to figure out at some point is 9F. When I'm winning machi krasso she pretends that she misses him so this one has an election inside. And I've checked with several speakers. So it's pretty reliable. But I haven't seen any other bad verbs or you haven't touched inside. But it's like inside. You can say as you're doing well I'm doing well, I'm not having it or I'm walking or you can say together I'm together, I'm having it I'm having it You can say as I'm 11G, I'm having it I'm having it I'm having it wrong. You can say I'm having it It's a kini komo ga mazu, it's a kini kiri kini komo, it's a more young person, but it's actually, it's a pop song, a rap. Yeah, you can even have a verb stem here for the result, for the bitimini, as in what's the term for one of the other j, kini pera mazu and quizians. That means that can mean the girl sang until she fell asleep, or that means that can mean the girl sang in her sleep. So here in this case, left edge element can be the manner, or direction of verb bell, or a melody, or a kind of linker that I'm going to talk about the linkers out there, or a nominal reference to the kind of song, or the verb stem, Tapa, 13. Tapa and drive someone, you can drive along, you can be Tapa and Arun Toten, along with someone. You can drive around, example of a song at the very beginning, the nipa-pama-tatana, drive around, go on, somebody around. You can drive here, you can drive without direction, you can drive from certain place, you can drive loud, isn't it? 13f, oma-tue-tapa-na-na-ma-shi-shan, it's for driving. You can drive slowly, you can be driving, it's in H, so you can have all kinds of letters here. So, directional manner verb bell is 13i, arun, kya-tapa-na-ha-ma-shi-shan, that's what it's called, it's one-tend alteration here, that's the left edge element that's applied, direction, manner, result, or complete of these things. Now, these are usually elements that are scientific components of event and position. So, I'm going to propose that the stamp, the orgy-cream, in particular, the background stamp corresponds to an event. And all the elements in the stamp contribute to the event and position. So, syntactically, we can represent this event phrase, pp, and the structure is on the next page. All that you need to notice is that the left edge element, so the epi is the left edge element here, gimme, and it originates as a component of the root, wii is a wii root here, originates as a component of the root, and then moves to the left edge as a, it looks kind of sticky, what is here on the movement. So, but I have evidence elsewhere that it moves. I don't have an explanation why it moves, but I have evidence that it moves. But it's all, it's irrelevant to focus on the purpose of this talk. Also, all you need to remember is that the gimme is inside the epi, it's the left edge element. The epi is the stamp. Now, I also do the event composition when it says syntactically, and can be visible through the scope of the verbose in particular. So, some of the verbose, higher level verbose, need to take scope over the event, but lower level verbose participate in the verbose. So, there are these two kind of verbose. The distinction is very crucial here. Now, if the proposal is the stamp is an event, then we predict that only those elements, only those verbose that participate in that composition can be able to appear in the left edge. The element that higher level shouldn't be able to appear in the left edge. But the epi takes scope over the entire map. So, this is the prediction. And now I'm going to test this prediction with the, so far I've looked at what can appear in terms of just over you. And now, what can't appear? So, the trailer shows a verbose modifier that I mentioned at the beginning, that we're seeing the stamp. They can lower into the stamp. We've only already seen an example of that. So, steamy, steamy bakto along outside, inside the stamp, it's right. But the same, the same, a verbose can appear outside. That's the question that I asked at the beginning. Why sometimes inside and outside? So, here's the structural representation of that. You can say minu nakusi. She looks nice, minu inside the stamp. Inside the, I'm sorry, minu here is outside the stamp, because nakusi can settle its own. So, in 50, you see that there's two UTV levels. So, minu is not, you can hear it here. Now, nakusi is a verbose. It's just outside the first UTV, so outside the stamp. You have to say minu nakusi, it smells nice. To have minu inside the stamp, nakusi is not a verbose. It's a root meaning stamp, a smell. So, you have to have, you have to smell nice. You can just smell it. Also, English is a different thing. Yeah, so this is, appears inside the first UTV, it's its structure, so it's its own left edge, it's a big trigger. Now, in this case, there's no obvious difference between the meaning of the verbose outside and inside the stamp. But the proposal predicts that, with other preverbal modifiers, there will be a difference, or some of them won't be able to lower. I'm going to test this with several kinds of verbose. So, higher-level verbose, like speaker-oriented sentence-level elements, then relative preverge, linkers, I'm going to tell you what they are, and a spectral element. One example of a higher-level element is mochi. Mochi means just the speaker-oriented element. Mochi, asma febi mati, just sitting in the sun over there. So here, mochi is outside the stamp. But, so we predict that mochi shouldn't be able to appear in the left edge, it's outside the stamp. But here, you have a minimal pair of 18 A and B. So you can say, here we are, I'm your argument, nationality, mochi image. I'm going to talk to you about it just the other time, mochi image, mochi is outside the stamp, it's fine to say that. But you cannot say the same thing using the verb mochi shimochi. Shimo is a root meaning dance, it needs something like that to actually become a stamp. You can't use mochi to complete the stamp composition. So the 18 B is a grammatical mochi, you can't appear in the left edge of the stamp. I'm going to skip the rest of this and move to 2.3 questions. So, what do I call linkers? They're also both relative groups, relative reverbs, non-confirm literatures. So this element said link the verb to some accompanist to indicate outside the verbal context. So in a certain place, and then the place is indicated somewhere or in a certain manner. So onchi in particular means from a certain place or for a certain reason. But it can have several different meanings. So it can be directional and this is meaning as a pre-verb outside the stamp. You can say shabana mochi dake and a cold wind blowing from the south, onchi outside the stamp. It can be encoded, it can be this 22 B, kata-tabe-onchi-tapisite, if you've got cold something there. It can also form negative facts, because it's 23, onchi-tipi-tapisite, kama-tipi-tapisite, so that means that everybody believes in each other. Kama-tipi-tapisite. Now, I also predict that only the directional meaning of onchi should be able to appear outside the stamp because that's the only one that contributes to that. And this is indeed the case. Directional onchi, grammatical outside the stamp, you can say shabana onchi no weh, onchi inside the stamp, you know the left edge, which is going from the south to the end, it's a minimal variable, 21 shabana onchi dake no weh, onchi outside the stamp, but shabana onchi no weh is inside the stamp, it's violent inside. But important, onchi is grammatical stamp, so you can say kata-tabe-onchi no weh, something like a wind blowing, that is a, you cannot, you also, so when it's all actually fixed, try to fix it by a certain location, so kata-tabe-onchi no weh, you want the wind to go from there, you want to be, to make it directional, so it can be both of them. Now, onchi no weh, it's past also grammatical inside the stamp, so you can say, kaonchi kawe duapus, there we didn't give any tax, that's grammatical, it has to say kaonchi pime kawe. Another linker is the ishen, means to a certain directional place. And it can also have several meanings inside the stamp, outside the stamp, and also pime. So it can be directional, weh ishen pime kawe wapus, the rabbit's tracks are going in that direction, to pime kawe's stamp, that ish attaches to it. It can link the vent to a temporal antecedent, as in 29 A and B, so you can say, meh meh guach eki sih seh pastiyan, in thai jhene kawe, I often sing in the shan or so, so nika moe ishen meaning sing, ishen links the stamp to the external circumstance, while in the shower. It can also form an event nominal, as in 30, so iwi jhene kawe ishen kawe ish pastiyan anima, so together with the suffix k here, in the common test count, ishen forms the vent on the crack session. So again, we predict that only the directional eji should be able to appear inside, but not the other two meanings, and this is indeed the case. So you can say, 31, wait eji kawe wapus, rabbit's tracks are going in that direction, and that's fine, just like 28, wait eji pime kawe, directional eji inside is fine, but event-related eji ishen kawe ishen inside the stamp. You cannot say meh guach eki sih seh pastiyan, that eji hama ishen, where hama ishen will be in sing, we are trying here to put eji inside the stamp, on the left edge, that's agro-micro, we can say eji niha masu. So meh guach kawe puke sih kawe wapus, during the winter rabbit's tracks, so the intended meaning is temporal here, during the winter, in temporal antecedent, but the only possible meaning is directional, during the winter the rabbit's tracks go over there. And the event nominal eji is also agro-micro, so you can say eji kawe sih pastiyan, kawe wapus niha wapus, so you can have eji inside the stamp there. So as predicted by the proposal that the stamp would event, only the directional meanings of linkers can appear inside the stamp, but a spectral, a negative past, and event-related meanings cannot appear inside. Now let's look at the directional and spectral, there's restrictions on a spectral at the elements. So spectral elements refer to, in general, refer to the internal structure of the event that is noted by the group. Now, the error suggests that a spectral modification cannot appear in the event internal there, but has to combine with the core event. There's also another view that it's present any, that spectral at works can appear above the event level and inside the event level. So there's two different kinds of spectral at work. So a spectral at works of the core event level can take, can take scope over the core event, or spectral elements within the core event must participate in the event. We'll see the difference between these two here. So I'm going to look at three spectral elements, key web. Again, we'll back. My chief start and Pune stop. So here's an example of key web back. So for instance, we have this scenario. A mother is waiting for a child to finish playing and come over. As a child is about to run towards her, the child yells, I'm running back already. So in this case, you can say key web outside the stem, key web impact over, that's fine. You can also say, I'm going to give a talk to give inside the stem. So for directional meaning, both positions of key web are okay. But here's another scenario. If for instance, I had sorenko, which provided me for following my running routine and when I finally got healed and I go for my first run, I yell, I'm running again. In this case, I can say, I can say, I can say, I can say, a key web impact over, key web outside stem. That's it, again, that's the special again, right? But I can say, a key web active, maybe directional. And another example, this is a very first. So I can say, key web outside the tree is going again, key web outside, but I can say, key web ishpope, key web ishpope, but it sounds like, I can solve it, It sounds like a tree is going backwards. So I'm going. So when the keyway is inside the stamp, let it in all the time direction. You need to not have that effect on it again. Another example is with Machi being start or wait. Now we can see that Machi behaves differently with motion verbs and non-motion verbs. So motion verbs is in the drive. Only the directional every line of Machi is a bit of a temperature. So this example, for example, I'm going to start driving next month. Machi is outside the stamp. For example, I'm going to start driving next month. I'm going, it can only mean I'm going on my own trip. So in the second example, Machi is inside the stamp. Machi is left-handed. It can only be directional. I'm going over there, driving away. It cannot mean I'm going to start. It's a non-motion verb and there are two different strategies. So some speakers depend on a speaker, which strategies are void. So for some speakers, Machi, inside the stamp ends the directional component instead of the spectral. So the verb angry, the angry is not a motion verb. So you can say, haja, Machi, ishivazi. Machi outside the stamp, it just means start to get angry. You start to get angry. But if you're trying to put Machi inside the stamp, with the meaning angry, haja, Machi, na waise. It means something like she's storming off anger. She has a movement of point. Angry, right away angry. Now a second strategy for speakers is quite interesting. So Machi can't have a spectral reading inside the stamp of a speaker, but it can only have certain restrictive flavors. So for instance, if you just grab an episodic event, then Machi can appear either outside or outside the stamp. So you can, if you want to say, my baby just started crying right now over there. You can say, haja, na feva, machi, maté, macho, demo, started crying, mat, in this case, inside the stamp. Or you can say, maté, machi, mawe, started crying outside the stamp. Either way, it's fine. If you're talking about an habitual event of crying, then you can only have Machi outside the stamp. You cannot have it in a blood test position. So here's an example. When my son was a baby, he hardly ever cried, but when he turned two, he started crying a lot. So we're talking about the habitual crying. So in this case, you can say, it's a machi, mawe, it started crying a lot, machi, as the stamp, it started crying right now. You cannot say it's a machi demo, that means it started crying right now. So the stamp, external machi, takes place in the event. They see the stamp in the event, and the stamp, external machi, takes hope over it. That's why both readings are possible. You can have, start, inside, demo, outside. But if, for instance, you stop eating meat completely, and that's in 46, then we're talking about habitual non-eat meat, right? You became vegetarian. In this case, 46, and you can say, you keep eating meat in Uyaz, I stop eating meat, with Puni outside, but you cannot say, Niki Pun and Uyaz, with Puni outside of the stamp. Just like with the machi start, if you're talking about it, one time it's fine to have the stamp inside the stamp, you can contribute to the event composition, but if it's an habitual event, you have to have an event first, before you can, because spectral events cannot appear. So it's a little page 23, so this is all the restitution that I want to look at. So we've seen that higher-level elements like speaker-oriented elements cannot appear inside the stamp. Spectral elements have only a structured range of flavors inside the stamp, and linkers also can have meanings like directional, or manner, but they cannot be event-oriented. So hopefully this convinces you that the stamp, this language, because of the metaphosal of the stamp, this language is an event, and all the elements of the stamp contribute to the event composition, okay? So to answer the question, what was the letter, shall I not do it, in biker terms, the composition of the stamp? This also shows you that the structure of the stamp cannot be explained to users. I think that I've been working on it more recently, but it's part of the same larger project, so the darkest days is the stamp that we looked at at the very beginning, the stamp has all three elements of the template. Darky, initial, sit, medial, and final, but it's actually a school fee, so it has all three elements of the template. Now, again, we want to analyze the stamp beyond the traditional template, and again, I'm going to ask the same question here. So if we have, if the suffix, if the final means the non-possession half, if you have the, if you have the, you know, why can't you just say sit down? She has feet, is that 50? It's ungrammatic, we'll have to, she has to have cold feet in all the parts of the agreement. So it's also the question of the left edge requirement, what is the left edge, shall I not do it here, but I'm going to answer it in a different way. So why is it, in this case, we have to have to say that she has cold feet on this day? Why can't I just say she has feet? There's another way to say she has feet, she doesn't have to have cold feet. Yeah, so basically what is the morphosontactic makeup of this stamp, and what is the relation with this stamp? I'm going to propose, eventually, that the suffix, it doesn't really be a non-possession. What it does is that it selects for a small clause complement and relay the proposition of a small clause to the enemy argument in the specified. And the edge of travel element is actually the very good one. So first I'm going to look at some properties of possessive stamp, because it's also a very under-described, under-described construction of the monitorism, and it's very, very reproductive, so. You can have the different types of nouns and modifiers, so you can have, here's some examples in 3.1.1. Mishinomabete, of course, is more sometimes the many teeth, Mishinomabete, many teeth have. You can have it with all kinds of ineligible possess nouns, like teeth, pants, blankets, so party parts, and co-designing ineligible selves in this language, relatives, like daughter and partner. You can have coffee, you can have a new partner, you just can move them. Any kind of nouns are also ineligible possessed, as in 52, you can have, you can say, the kipakunem, which is moniken, for your modifiers for health care, you can say, in tokski, or kikindasne, have a new telephone number, new telephone number. Let's go to vocation number seven. Noun can also be incorporated with one of our most popular, it's too hot, so you can say, you can incorporate the okski, the first one, the okski, the first one, the okski, the first one is the okski, the first one, you can say I have two big heads, incorporating the noun, big head, nijomakitiguane. So, now, here, you can also be referential. The moniker has two heads, with the noun added appropriate, and then you can refer back, you can say one is big and one is small. And you can also have modifiers for this noun, outside of the verb stem. So, you can say, nijaktiguane, the okski, the moniker has two heads. And then, pejik pakasitiguane, and konin pejik manitiguane, where in this case, the number of pejik is outside, outside the verb stem in the sentence. It's not a nice to the noun, to the noun. It's trying to modify, okay? So, these are all properties that are common. There's noun-properation instructions in a language, and have been noticed for other non-properation instructions in the juba, in the jubu particular, but they weren't described in this particular construction. But the most interesting question is, what is the structure of the sentence? Now, to repeat the celebrated that happened before, is why can you say the is it there, but not sit there? 159a is it if, but 16 is not. You have to have that part, that whole thing. Now, to answer this question, we're going to look at some data, some examples that have to do with the verbs. So, let's look at 61. Is shiwabu biwabu okay? So, this verb is, looks like it's well-formed verb, in two-degree articles. So, you have a modifier noun, a corporate noun, and then the category-fine itself is there. Shiwabu biwabu is there, I have a verb cap. Shiwabu is there. It has the old fashion. It doesn't have shiwabu. It's true, they say it's so. For some reason, I'm grammatical, but if you add the modifier new, as in 61, that's fine. We've got shiwabu biwabu again, I have a new verb. It's fine. So, it looks like you have to have another modifier here to make the stem grammatical. Here's another example. If the commissary grammatical has a pencil, this sort of pencil is a point, but it's a modifier. We have to say, I have a new verb, a pencil. You have to add something else first. If you make a weapon in Kalpana, it's bad. You have a toy car, and you have to add an old component, but if it's a new toy car, it's fine. Now, why is that? Because it follows immediately. If this is a grammatical example, this example that we just showed, a modifier, if they used in a different construction, so if you focus use something to make the letter to another post, then suddenly they're fine. So you can say, I have a pencil, maybe commissary grammatical, but you can say, miktigum is a grammatical, shiwabu jeneth, where shi is in a pattern particle. In this case, it means she has a wooden, she has a pencil, or she focuses on the miktigum wooden part, she has a wooden part, suddenly it's fine. Here's an example of a toy car, so if the equations can't jump to the airport, I say, kawali mitababin tabin etha, in this case, etha only focuses on the toy part, they all have a toy car, and then in this case, I have a toy car, it's fine. And there's more examples like this, nipakunesic alakuna means, kakunesic alakuna means the donut, so it's literally a bang that has a hole in it, but it's a donut. So banak is, you know what banak is? Yeah, it's like a traditional flat, or something that's not flat, deep-fried, some of the big flat that people make. But yeah, so kakunesic alakuna means donut, literally banak with a hole in it. But if you say something like nipakunesic alakuna, look, it can only mean my banak has a hole in it, it can mean I have a donut. So this varies from one speaker to the other, so the sound speaker is a small kaki-sine, she shoes a big, okay, the sound speaker is not okay, the sound speaker can only use it under focus, so the wasamsa nipakaki-sine, much of that too big. So I'm going to propose that the suffix e takes a small pose component and then relates the position of the small pose to the animate argument that's introduced in the specify. And it's something that's been proposed by the verb half in English when it's said on some authors. So the structure is in 67, because again, the structure of the small pose would be a different thing. The experiment that I was changing, basically, it takes a small pose component as specify, small pose in this case is car new. The subject of the small pose, new or she's a particular small pose. And what the suffix does is the proposition, car new is true of this animate argument. When you assume that the ocean is predicated, then it derives the focus effect that we discussed earlier because the subject are all principles, principles of formation. And predicate is a new information. Car is new as opposed to, yeah, just being a modifier, new car is predicate here, car new. In terms of the possessive, possessive now preparation also look more, they're more complex than they look. And they're very, very dynamic structures. We've seen that they're very, very productive. And they actually involve, are we able to have a small pose of preparation other than traditional and non-comparison in the traditional sense. So the large conclusion for all of this is that I've kind of combined with the course that I've been working on. Yeah, I want to give you an overview of how complex this stuff is and how it works. And that you can't use a template to describe any positive language in all degree. Analyze each type of template in particular, separately and we still wonder what award is in this language. I'm a vicious dog. You lost me a little bit towards the end, I have to admit, but potentially because I was really interested in what you were saying about the linguistic coding of events. And I mean, you just said in your conclusion that you really have to look at, you know, every construction in a real life, but I would still like to ask you about possible generalizations you may want to make about classes, about stems maybe. I mean, some of them, just glimpsing and then look as if they're always transitive or they're always an object, which might also be an accessory that's used today. In any ways, you know, between parallels between actual arguments and piece, or some look as if they always need to have a manner component and then event structure. So did you attempt any generalizations in terms of event structures, predicate decomposition, or looking for parent constructions and automizers in the band? You mean the operators always go over the entire events, segmenting them into sub-events and trying to form a life. So let's you do that at some point, but I feel that you need to look. Oh, you're not necessarily formal. I mean, but just as a semantic termization, then maybe you can enter it. Yeah, there's different classes, the types of, like, if you're talking about vibrate types, there's some types of that. So there's a type that takes everything and that it's very hard to distinguish. What is the difference between a loop that means sync? Yeah, in this case, it's a kind of challenge for me because it looks like it's arbitrary. So it's saying, you know, it's just an arbitrary. In some cases, it's, you can only have the only manner of work for it. Yeah, well, only the sample's different in per. Sorry, I do have a very small question. In example, 13K, I'm just not too sure what one of the elements in the graph will be. So I just wanted to check for you, please. In the 30K that you said they are. So what is that one, three more than one signal? I'm not sure. So this means that it's a transitive or a perfection. It's a first person, actually a third person. First person is the subject that the person is on. That's what it is. Okay. And similar in the graph. Yeah, there's a lot of. So in 9F, we have both first person and the sustained loss. Yeah. But they're both third person. You mean the man who put him in the math class or she put him in the thesis class? Yeah, it's separate class. The first person I could solve. But they're both third. They're both third person. They're both third person and missed one. Oh, that's true. That's what wasn't there. I wasn't able to show why I had that. So it's supposed to be in the graph, it's supposed to be a version of it. I pretend I missed it. But if it's third person subject and third person object, it will be different. Yeah, yeah. I think that's how you would say it. Okay. Yeah, also you can have over things like that. I don't know what you call them like third. So I don't need to go into the final steps. They've got animate and transitive. But they're different from different verbs. Like is that some sort of parallel? So you've got in 2A, it codes animate and transitive. But then you have a zero morphine for that in 3A. It's a closed class, but there's... Uh-huh. There are several verbs that indicate that the verb is transitive. So if you get the verb transitive, sometimes they have additional lexical meaning like wrong or something like that. Okay. So they mean over things, otherwise. Sometimes they have... sometimes they indicate that it's also like an underdog question. Right. Sometimes they indicate that the type of argument is a single argument. They're asking data questions. Like along the same lines, I wasn't always clear what marked the aspect or the tense. So the tense is marked by the outside tense. So key, like the one in the page. In the past ones, yeah. But so for instance in like the examples of 3... You mean the last one? I mean whatever it is. There's no... So like 3A versus 3C, how do you know it's not she is dancing well or 3A, 3C, 3C, she dances well versus she is right. So other way. I don't know why I would want to do it differently, but it could be she dances or she dances. So what indicates? Oh, nothing. We don't know if you can use it other way. You can say she's dancing or she dances. So that's like a sort of default for verse 3 and then the past would be marked with that key. And then also along the same lines, what indicates it's the third person singular? The last person so again it's like a default. So this is like the base that we can get. Alright, that's it. Good question maybe. Your analysis of possession, as I say here the proposition developed by the small cause the eduement argument how important it is. Why does it have to be and what actually follows from the setting? Sorry. I guess you can't use this verb of animate such. You can have the table as the corner. Pro-typical possessor. Possessor is not an idea. No, it's a grammatical distinction. The verb, I mean the nouns are either animate or animate. Basically for your analysis for the structure. Or tensions. It's not the last possible. We have to thank the speaker once again. Yeah, I'll do it in the box to get myself on now. Yeah, maybe you'll leave the people who do not know about this. Because I have new students. Okay, sure. So if anyone who wants to meet chance to regroup and then in 10 minutes I can meet you just outside the front. Steps? That's no obligation.