 but of course, he's a PhD student in the physics department. And my special job today, and my pleasure, is to introduce our home ground speaker, Dr. Yen Jiang. Dr. Jiang has more than 20 years of academic teaching and research experience. He moved from Hong Kong Polytechnic University to SOAS last year. But actually, he has a long-term relationship with SOAS and the University of London in general. He graduated with PhD in linguistics from UCL and SOAS in 1995. And he's also an editor and contributor of SOAS Working Papers in linguistics and phonetics in 1993, which is now called SOAS Working Paper in linguistics. And he has more than 60 publications, both in Chinese and English. And his research focuses on formal semantics competitive semantics, pragmatic, and rhetoric among many others. So I'm sure we all enjoy getting to know his topic much as you do about bill in Mandarin Chinese. OK, please join me in welcoming Yen. Thank you very much. Thank you for Ms. Lulu for her very kind introduction. And it's really a great pleasure to come to the departmental seminar to give a presentation. So I have prepared a rather detailed handout with glosses and romanization for Chinese sentences. So I hope that it will make easier for those of you who don't specifically work on the study of Chinese syntax. And so I am prepared to go through the example sentences very closely together with you so that you can see the point of this topic. I title it as much about dough for a special reason because this dough is very much like the English every, or the English all, or the English even. And that, strangely speaking, has attracted quite a lot of attention among the Chinese traditional and contemporary syntactic circles. It seems that everyone wants to have a dip of finger in the topic. And every time the theory or the trendy theories change, then people want to do the topic again. So that is why there has been a steady accumulation of literature on this topic, around 200 journal papers or book chapters. And many people got their PhD degrees for working on dough, whether in English or in Chinese. But then the majority of work have been done in Chinese language. And I think that there is even some kind of gap between people working on Chinese from the English literature and people working on Chinese because the data is rich and the native speakers' intuition is important in making judgments to the extent that journal papers published in English tend not to quote a lot of Chinese literature. Giving people probably the false impression is that the authors thought up these ideas by themselves. I think that this is an unwanted outcome, as far as I can see. So I will introduce the background and part of the earlier work I did. And then if we have time, and then we talk about some recent papers on dough by some of the other colleagues and peers. I think the study of Chinese syntax is complicated in its own sense. One is that Chinese, as is well known, Chinese lacks a rich morphology. So morphological and inflectional features are almost new when we study Chinese grammar or Chinese syntax. That makes it very different from the study of syntax in romance languages. On the other hand, then Chinese grammarians have to rely a lot on the distribution. So there can be a lot of permutation. And then people result to the abstract theories on the analysis of language, for example, move WH on a language that doesn't have overt WH. Or providing a case theoretic description of Chinese in which no explicit case inflections are shown at all. But people manage to work on it. And there is also the added complexity for Chinese in the sense that politically speaking, well, only allowed, for example, in China to talk about just one Chinese language, which is Mandarin. And all the other variations are considered to be dialectal variations. But in fact, what is very prominent from phonetic point of view is that many dialects are indistinguishable from one another. But the study of syntax is very much suppressed in the past until only very recent times. For example, the study of Shahini's grammar has been very much suppressed to the extent that there is only one known Shahini's grammar written by the contemporary scholars. But before him, then it was the missionaries who wrote all the interesting Shahini's grammar in French and in English 100 years ago. But then now we can try to look at these studies again. But in the past, the results have been very much ignored in the field. And then there is also the non-homogeneous ingredients in Mandarin Chinese if we're talking about the very northern ones or the southern ones. So it's not a homogeneous language we're talking about. And finally, there are all these constraints because of the characters we use. So that makes us oblivious of any possible inflection or suffixational particles that may exist if we were to transcribe the language from the initial stage. But now we tend to ignore it because we only look at the characters and the characters don't recognize these. This is especially the problem when we look at dialects. Well, those dialects will have to follow the formal conventions of the Chinese characters without really showing their individual characteristics. But so I have given some summaries of the idiosyncrasies of Mandarin Chinese in the first page. But I don't have time to talk about them in detail. And today I talk about a special problem which is related to the adverb of scope or adverb of quantification. In Chinese there is a group of about a dozen such adverbs of scope that command our attention. And each of them is special in the sense that its sense can be polysemous in the sense that we don't just have one single unique meaning for each particle. But then for these adverbs of quantification, the context usually decides which interpretation we can obtain from it. But then on the other hand, these senses are also very logical in the sense that they are probably the nearest to what we want to characterize in terms of classical first-order logic or in some modern variations like formal semantic representations. And they are also rather different from the encoded meaning of related English adverbs. So for example, in English there is every or and then even. And then not many people would like to associate even with or. But in Chinese, then they happen to be of one encoded word or particle. Well, in general, this is not too idiosyncratic because if we look at many European languages, then we do see some similarities. And that has been worked on by some people that I'll mention later. So the problem I will talk about today is to use it in a rhetorical way, how to get even in Chinese. I talk about the data first because it will require us to get ourselves familiar, right, with the data. So the first is how Chinese expresses convocation, like if we use the first-order logic, then we know that there are ways to express universal convocation and its existential convocation. But what we care about here is mainly the universal convocation, which is on page 1, the last line. So if you want to say something like, every boy sleeps, right, we use the logical formula as OX, it was a missing X here, OX, boy X, and sleep X, something like this. And then for English, then there are several ways of saying that every boy sleeps or all boys sleep. And on page 2, there is the existential variation that doesn't have to concern us here. But page example 3 gives us an example in Chinese, which is special. It is different from English in the sense that Chinese has a rich, classified system, right? So the expression of universal quantification, like every, can be reached just by the repetition of classifiers. And that, in fact, was the older, the more colloquial, more oral way of expressing the idea. And that was also recorded in earlier works of Yuan Ren Chao in his paper notes on Chinese language and logic. So it can be something like, nǎng hǎo ge ge du sui le jiao, right? So we have the gloss here, boy and ge ge, classified. The fact that I repeat the same classified means universal quantification for Chinese. But that is one possibility. And note here that in example 3, we already have this use of do, d, o, w, with the first tone. So it's do here. So it seems that the mere repetition of classifiers is a sufficient condition. But then, at the same time, we also need do to be present. So it sounds like boy, boy, or sleep, right? Boy, boy, or sleep. So it looks a little bit redundant, right? Like you have boy, boy to talk about universal quantification but why you have all at the same time. And that is a question that people talk about later in their studies. But then there is also the more common expression of universal quantification in example 4. Mei ge nǎng hǎo do sui le jiao. That's every classifier and boy and the all sleep. And this, I believe, is a later kind of expression, partially Westernized expression. But then in modern Chinese, this becomes more commonly seen, especially in written language. And it is this kind of sentence that commands our attention when we deal with do. So do is strange in the sense that first, it can be translated as all, right? And it expresses universal quantification. But it seems that on the other hand, we also have this each classifier and plus now construction, right? So it seems that universal quantification is achieved by these two means at the same time, right? If we take away this do, then the sentence is not considered to be well-formed. So if we have every boy sleep, it's not enough. You need this do support. So you have to say in Chinese, every boy all sleep otherwise and people won't accept it. There are some exceptions, but we don't have time to talk about it, right? But this is the canonical form. And so compositionally, this presents a puzzle, right? Why do we have these two bits and pieces? What is the compositional contribution of each? I think that people working on the problems do don't come up with a satisfactory answer. So here, many questions are probably still open, right? And because people don't have ready answers, then a lot of people have to come up with some ad hoc stipulations that I think is the state of the art. We try to avoid these arbitrary stipulations, but then people don't really have commonly agreed solution that is satisfactory. So what is the contribution of this each, right? If it were a universally quantified structure, then why do we have dough, right? But do we have both of them contributing to quantification? In that sense, then it will be at odds with the stipulation of logic. Because in logic, then you tend to think that only one unit will serve the function of universal quantification. At least, we don't have a compositionally sufficient analysis for each of them. And then there is the 2.2 is a question that I started discussing about 20 years ago first, is that we want to know what is the quantificational target of this dough or all, right? So if we take this dough as something like a floating quantifier, then what does it quantify? So what is this? We call dough traditionally as a quantifier of scope or adverb of scope. So what is in its scope? What does it work on? And we opened up this discussion topic about 20 years ago, arguing that its scope, its quantificational scope, is always to its left. And that gave rise to a lot of follow-up discussions. But why only to the left? Well, it's because when we studied the data, it seems that Chinese simply works like this. And it's not too surprising because if we look at the NP structure in Chinese, then we know that the NP structure is strictly head final. And it's just typologically works like this. So we argue that the target of dough is always or only to the left. So we look at some examples. For example, in 6, Tzu Shi Su, this book, he and dough read, and then the aspect optional suffix. So it means that he has read all these books. So translation doesn't reflect the use of this dough here. So you have to say he or read these books. And this all is related to these books in the topic position. So it has to be plural, generally, for dough to be related to the noun phrase. So it's to the left. And 7 is an example where you have a NP between the subject and the predicate. This is now taken to be grammatically the grammatical focus construction, what appears between the subject and the predicate. So something like Zhang Shan, Zhe Liang, Zhe Liang, Meng Ke, all showed up. So something like Zhang, these two courses has taken. But in the universally quantified construction, you have to say Zhang, these two courses all has taken or all has taken. There is no special word for both. And people may also wonder whether this dough here works as a universal quantifier or not, because it is simply a matter of plural NP. If it's not universal, at least, it works like an exhaustive quantifier. So again, the bolded unit is to the left of dough. And number eight is the typical Ba construction. So Ba used to be a content verb, that is a verb with a concept in classical Chinese meaning But then the meaning of Ba was cementically bleached over the change of time and the change of language. So now it is used to add a noun phrase before the predicate. Some people would call it a causative construction. But the causative construction, if taken in this way, has a lot of non-causative meaning. So some people would also call it the dispensative or dispensary construction. But from the recent talk we organized last week, we also learned that some other people would argue that this Ba construction simply functions as promoting one post-verbal NP to the front, because there seems to be a very general vague and again, inexplicable phenomenon for Chinese is that the language hardly tolerates two linguistic elements after the verb, two visible linguistic elements, whether they are arguments or adjuncts. It simply doesn't like it. There are some exceptions. But then the further north you go in China, the fewer cases you will find for sentences containing two linguistic units after the verb. And this is considered by many people as a strange case. But this is also the special typology for Chinese. And some people even put it in the form of an X-bar convention. So if you are talking about this X-bar convention we have to be careful, because other people would mistake you to be talking about X-bar syntax in general. But it's two very different things. This is descriptive case. So number eight is a sentence in which we have this special Ba construction. So it's like Zhang and this book but headed by this Ba construction. It's like proposing or attracting an MP or some other element away from the post-verbal slot. So it's like this Zhang, Ba, every book. And this Dou read once. Dou read once. So it's like he gave a reading to every book. So again we see that Dou quantifies to its left not to its right. And then nine and 10 are like this as well. But in nine we have the adverbial every day. Time adverbial every day. So it's not in an argument position. But it's also for Dou to quantify over. And then 10. So he at the beach and Dou take photos. So that means that he took photos at every point in the beach. Then the controversy was on some apparent counter examples. And that was what many papers discussed on, especially works in the 1980s. So I have included a very detailed bibliography, thematic bibliography containing about one-third of the papers and works on this Dou. But there are many others. But these are the ones that I think have attracted more attention. So mainly in the bibliography, the works that I cited in the 1980s, they talked a lot about these descriptive counter examples. And 11 has now become a very famous example almost every paper tried to deal with. So it's like Xiaoli can probably be translated into English as something like Johnny. Johnny and all or Dou buy a tweet clothes. So for this construction, then it seems that Dou's quantifiable target will have to be to its right. Because that's what the linear order tells us. But the problem is that when we do a meaning analysis, we find that the sentence doesn't mean that he bought all the tweet clothes. What is meant is only that of all the things he bought, they are tweet clothes. So there is this difference between the two meanings. So it seems that the target of Dou for this sentence is still not to its right but to its left. Or at least, it's not to its right. It is something in the presupposition of the sentence. It's like of all the things he bought, they are all tweet clothes. But the sentence in the context does not contain all the information. And that is typical of Chinese. And that is the special characteristics of doing Chinese syntax is that all the time you want to talk about the syntax-paramethics interactions, because there is no morphology. So morphologically, you can't find evidence. So you change the context, and you change the distribution and see what evidence you can get. And then for, so this is what we propose to reanalyze example 11 into something that will argue that the quantificational target of Dou is not to its right, but it's something that you don't see in the presupposition. And for example 12, that is another kind of sentence. So if we look at the explicit sentence content, it seems that those quantificational target can only be to its right. He doesn't eat others or eat steamed bread. In my talk, there are a lot of examples concerning steamed bread. I hope you're not unfamiliar with it, or I'll get some chance to make some for you in the future. But it's all about steamed bread. So it's like, well, he doesn't eat anything else. Dou eats steamed bread. So it seems that Dou has to quantify over steamed bread. But then logically speaking, it doesn't make the representation well formed if we were to translate it in this way, because you can't really say of all the steamed bread that he had it all. Because again, it carries a presupposition that he has eaten something. And then of these something, all of them are steamed bread. So again, we argued for reanalysis and through which Dou quantifies something invisible, not to its right, but to something probably hidden in the presupposition. But we can't really say it quantifies to the left either, because what is presupposed is not explicitly stated. We don't know where it is. And maybe we accommodate it at the flesh of the thought when we process this, didn't eat anything else. So with that, we accommodate the presupposition. And 13, he all wrote novels. So again, it doesn't mean that he's the only person that created all the novels in the specific domain. What it means is that of all the things he wrote, there are all novels, not anything else like poetry. So again, we argue that it is something presupposed that is quantified over by Dou, but not the rather generic noun, noun NP to the right. But this is only the type 1 cases that were raised by the more senior generation of descriptive grammarians. And there are also the type 2 cases. The type 2 cases are also tricky in the sense that we have the nouns in plural form appearing to the right of Dou, but then still it seems that we don't want to take them to be the target of the quantification force of Dou. Because take 14, for example, I or inform them. So logically speaking, it seems that it is quite OK to say that I have informed them all in Chinese. But then the tricky point is that we have to understand them as anaphoric. So it's not OK if we replace the anaphora with the real names, like in 15. 我都通知小王,小李,小趙了. You say, I all notify John, Peter, Mary. That doesn't work. Because although what follows Dou contains three nouns, we don't usually put it in this way. So I try to reanalyze the type 2 cases as involving anaphoric cases. So what Dou quantifies over is not really the anaphora, but the antecedent. And the third cases involve WH expressions I can omit because one puzzle that many people reached is that this all doesn't seem to relate to a WH word. I think it also is the case in English, because there is the awkwardness of saying that what did you all buy, meaning that this what is plural. It seems that you're just about to ask the content of the what. And so it seems to give us a processing conflict when you say, what did you all buy. But this all doesn't modify you, but is used to modify what. So I omit the detailed description of that. But if we look at example 19 on page 4, I want to know who bought what. Who bought what. That will give us usually a pairing list reading, like in A. But then it is also likely that for each person, he or she bought a list of other things. So it makes the option open if we use 19. I want to know who bought what. So here no Dou is involved. So I think that for 19, we can have both 20A and 20B as representations. But for 20, if you say, I want to know who Dou bought what. The difference, I think, is that with 20, you presuppose that everyone bought a multiple list of things. So for 20, then only B is available, but not A. But the reason we want to make this minimal contrast is that we want to show that Dou is related, again, to something presupposed, but not to this WH word to the right. And for later discussions, I didn't follow the topic with too many writings. But then if you take a look at the bibliography, then it's number 57, 58, 59. The person who works, Professor Yuan, working at Beijing University, he has written several papers in support of this argument, saying that Dou never quantifies a WH word, even if you want to take the WH word as plural. And then I summarized the point again in this number 27 paper. So these are the three cases involving the apparent rightward quantification of Dou that if we can dispense with these cases, then we can argue that Dou never quantifies to the right. So this creates a special theoretical problem or research issue within the study of Dou is that quite some papers explicitly address this on the rightward quantificational possibility of Dou or on the impossibility of Dou in doing rightward quantification. Some people would argue that because minimalist syntax never rule out this possibility, therefore this is not the case. And they went on doing minimalist syntax on it. But I think that, well, at least we have to look at the data first. And this is still a heated point, because in the end, I will cite a recent paper that talks about this topic again. But then the study of Dou went on because Dou is related to the plural. It relates to universal quantification. So some papers started to dig out a lot of properties in plural semantics. And their description of Dou proliferated because they consider many plural features to be the unique features of Dou quantification. Or again, it's because plural semantics is relatively complicated. So no thorough treatment has been given from the point of view of plural semantics. But if we link this kind of study with the study and works of Goddard Link, who wrote a series of papers and put some plural semantics, then we know that many things match each other quite nicely. For example, for 21, which is again, which is another example involving steamed bread. So he and this dispensary construction, he dispensing one steamed bread or eat. Superficially, it creates a conflict because we want to say that Dou quantifies on something plural. Then how can one steamed bread be quantified over by Dou? But then we know that it relates to the possibility of a steamed bread being quantizable because if you have water and you have part of water, it is still water. But if you have a man and part of man, it's not a man anymore. But that is the case with steamed bread. If you have part of the steamed bread, that is still steamed bread. And when you talk about the predicate of eating, then you are really using the action of eating to quantize over the steamed bread. It depends on how you segment it. But then the action of eating will make the bread pluralizable in this way. So this is an interesting example showing that plurality requirement can be compromised or coerced when we look at the actual meaning. And here we want to say that context also provides the background. But then for 22, that is also an allowable example. So he and dispensing one bottle of water and all spilled. So how do you spill a bottle of water? Well, we can imagine you spill it bit by bit. So you can use some dough or ore to quantify over it. But let's suppose that if we produce a sentence like 28, no, 23, I'm sorry, 23, then he and bar a glass vase or throw away. Then it will be harder to process if we want to think about it from the point of view of universal quantification. The sentence makes sense from another point of view, that is the even interpretation that I'll talk about later. So it's like we can't really coerce a reading for it in which we segment part of the vase bit by bit because it's made of glass unless you smash it. But it doesn't make very natural sense to use dough to quantify over it just because practically it is not possible to segment it. So all these are related to the first foremost use of dough that is related to this universal quantification, which is equivalent to the English every construction. But then there is the second use of dough, which is even. And the more complete construction is a lian dough. That means some link plus an MP and then plus a dough and plus predicate. So if we say even Zhang came to the meeting, then in Chinese it would be link Zhang even or link Zhang dough come to the meeting. So it's like Zhang is the least possible one to come. So we say link or plus Zhang, like I'm adding the least likely item and followed by dough. And this is the second use of dough. And sometimes it is descriptively marked as dough two. The third one is the meaning equivalent to already. So you say dough 12 o'clock. Why don't we dismiss the meeting? So that meaning is like it is already 12 o'clock. So many textbooks and grammar books would at least give us these three uses of dough. Sometimes it's called dough one, dough two, and dough three. Descriptively, some textbooks would even come up with even more such cases. But then from a point of view of theoretical studies, we recall the famous crisis modified Occam's razor. And that typically studies these cases and come up with a unitary treatment. That is, we give a general and abstract enough definition of dough so that we have it as the basic meaning. And then all the other specific meaning is obtained through pragmatic inference or is obtained through the coercion of context. So I proposed a unitary treatment of dough. That is gradually turning the study of dough from a descriptive point of view to a theoretical treatment. So we propose that, in fact, there is only one dough or which is like universal quantification. It's only that the quantified scope contains nouns or a set that contains members that can either be unordered or they can be ordered in a way that if we think about it from the point of view of scalar model, then we know that some members in some sets can be quantified over, but then they have their own internal scales. The earliest studies on scales is probably Gilles Le Fauconnier's pragmatic scales. But then later we have a few more Paul K. and Goldberg giving us the notion of scalar model in construction grammar. So I was borrowing these ideas to propose that we can collapse three doughs into one. It's only that the target of quantification itself can involve a set which is gradable or a set which is unordered or a set which contains temporal points or intervals. So when we have temporal points or intervals, then we use dough to express this even lateness. So that will give us the already meaning. But if we have a scalar treatment of the target of dough as a set involving graded or scalar model, then we can treat dough as an even, carrying an even sense. But otherwise, then we have the unordered set which will give us universal quantification. So at that time, it created some differences to the study of dough because before that, there were people only talking about dough one, dough two, and dough three. And that is the typical traditional Chinese grammatical treatment is that they divide into different categories. But if they still find some discrepancies, then they divide them into further smaller categories. But our proposal was to treat it as a unitary one. There is one dough proposal. And there were a lot of follow-up studies. Some people talked about only two doughs were possible. Some other people still insisted that there should be three doughs. And some agreed that there could be one dough. So over the years, people worked a lot on it and hence the titles, much about the topic. So this briefly summarizes the work in these areas. But I would like to make use of some final minutes to talk about some recent works. So if we take a look at the bibliography, because most of them are Chinese, but that is what the state is. Page eight, I want to bring your attention specifically to some works. One is number 40. That's by St. George. Those of you who know him is that he was the head of the Linguistic Institute of Chinese Academy of Social Sciences. In China, position does matter. Well, so that's why I mean people attach great importance to his work. And his work comes with a rather surprising title, isn't it, for those who know. If you talk about this, the working out of the misleading study of dough, it seems to have some implications against some others. And indeed, if you read the paper, you'll find that he accused some others of committing the fallacy of dogmatism. And that, again, is a familiar notion in China because he was politically related to Marxist practices in the past. So I was feeling a bit surprised that he wanted to use these terms. So the whole paper sounds a bit authoritarian if you want to read it. But that is still not the rudest one. The rudest one is somewhere in the bibliography by a person called Li Wen-sun, year number 31. If you are interested, you can check out our Sewers Library database and dig out the paper and give it a reading. That's also an interesting paper. But then the other one is Professor Shulie Jing's paper that's number 52. He gave a lot of interesting cases to argue that, well, the Chinese dough is different from universal convocation. Because when you think about exhaustivity and when you think about the universal convocation, when you think about plurality, when you think of subjectivity, then it seems that dough always contains some exceptions. But I want to talk about St. John's paper a little bit more because he addresses the past literature and he comes up with a different conclusion. He wanted to say that the former studies suffer from the commitment to dogma, according to him, is that the study has been made unnecessarily complex. He argued that when you talk about universal convocation, when you talk about treating this dough as a maximal cover, that is, for some of you who know, that follows Swartzschild's book called Pluralities. Or for some people who treated dough as an introduction of a scholar-mised constant or something, that would infuriate the logicians quite a lot. But if you're interested, you can take a look at this so-called scholar-misation and the dough, which is saying in the bibliography page 7, number 9, universal convocation with scholar-misation as evidence from Chinese language, which is in fact a book I forgot to put on the publisher. I think it's Matthew, but I have the book if you want to take a look. So he was talking about these unnecessary, complicated cases, and he was arguing that the target of dough should be unanimously considered as rightward quantification. So that goes squarely against the early claims made by me and others. But when you look at the paper, then you'll find that he was not really talking about the counter examples, because what he adopted was a focused, theoretic, tripartite reanalysis of sentences involving dough. So I have included some excerpts from his paper towards the end on page 5 onward. So if you reanalyze a sentence into the focus construction that will start with the operator and then with the restriction part and then with the nucleus part. But if you do it in this way, then you distort the order of the sentence completely. Of course, if he wants to talk about rightward quantification, that's fine. But that is because he has carefully maneuvered the structure to suit his focused, theoretic reanalysis. So when he's talking about rightward and not rightward, it's no longer in the same sense as when we talked about it. And so that's the major thrust of his paper. And there have been people who are preparing papers against him, but then I have been reviewing some papers because they haven't been out. I'll have to leave it to another occasion to talk about them. So basically, I think the study of dough will still go on because people treat it in different ways. One recent PhD viva that I attended was recorded on page 9. It's number 68, as by now, she has already acquired her degree. So it's Dr. Zhang Ying, who wrote a dissertation on a typological approach to multifunctional adverbs in Chinese, which includes dough. But it's a dissertation done in Hong Kong University of Science and Technology using the rather new theory called semantic map theory that has been practiced by some typologists. So if you're interested, you can also ask me for that dissertation, which is extremely long because she wanted to do everything. Well, that's it then. Thank you very much. For his new analysis on this very mysterious field. So if you have any questions, it's a good time to ask. Or any comments, they're welcome. Oh, yes, Hannah. Yes, thank you very much for a very rich talk. In terms of data, it's outside of my area of familiarity. So I have lots of questions, actually. You mentioned early on a few times about the tension between the more oral and colloquial and then the more written, maybe formal style. And I wonder if you could say something about that and how it relates particularly to this. Are you seeing patterns with one tendency and written forms and other oral forms? Oh, yes. I think that the Mandarin Chinese has the way it is portrayed in grammar books appears as if it is homogeneous. And it is unanimously agreed by native speakers. But in fact, native speakers vary so much. For example, there is always the tension between people coming from Taiwan but working on syntax and people working in mainland China because they have been very sharp, harsh criticism on the data. And in the end, then those people from Taiwan will have to say that I have been working on a particular variant, which is called Taiwanese Mandarin or something. But then even in China, then there is also the tension between the more southern Mandarin, which is very much influenced by Cantonese, Wu dialects. Well, of course, if it's influenced by Cantonese, then it's not really considered to be authentic Mandarin. But if someone like me who is very much influenced by Shanghainese and others, then people can't really tell if I don't tell them that. But then the further north you go, the more special the structures would become. So I think that there can be a lot of disagreements on what you can say with those. Some people will put it in this way and other people won't agree. So a lot of papers listed here, they argue a lot about the data. Yes, please. I always felt that the Chinese was grammatically right, isn't it? Mm-hmm. Yes. Yes. Everything else was different, but the question was particularly easy. Yes. And so is this sort of Western influence the way or this analyzing it so meticulously? Yeah. Or are it the Chinese, or if you were in China, also analyze Mandarin so thoroughly historically? Well, yes. I think the, well, we say that modern linguistics came from the West. So before the first half of last century, then in China there was probably only, very safe to say, there was only something like philology. And the philology was studied to serve the need of reading classics. So, but then so was modern science, right? And Chinese, in the past, in China, there were only technicians and not real engineers and scientists. And technicians were not given very high prestigious places. But of course, one possibility is that the first emperor beheaded all those people who sought metaphorically, not metaphysically, right? And he killed about 400, 300 people. But then Chinese grammar was also uninteresting in one sense, is that it doesn't have rich morphology. So what I think a lot of conferences or classes when they look at romance languages, it's a lot of, well, for example, Irina's projects here, right? It concerns a lot of morphological variations that I think that they simply don't exist in Chinese. So that shows the variety of linguistics. But on the other hand, in China, people are many. For all the tea in China and for all the people in China. So whatever we study here, for each rather minor topic, then there are about half a hundred people working on it. So always there are quite a lot of people. And now, because there are so many universities, so many PhD programs, so these topics are no longer considered to be, well, too restricted for these people. And so what I'm talking about here, most of these topics are very familiar with by people over there in China. And with the added fact, there's many people studied here, and then they're now working there. So they're teaching about something from South West or something. And they said, usually for those people, they left from China by all quiet and meek. And after they returned from source, for example, studying government phonology, they became very aggressive. Yes, yes. Thanks. I think around, for example, 11, you have a kind of paraphrase of it, like all the things you bought. There is three clothes. And then you also talked about this, though, like two, which was the least likely. Yes, the even reading. Even reading. And I was thinking that both of those really draw heavily on real world knowledge and expectations. So you need to know what is expected in order to be at least likely. So yeah, it's interesting that kind of, yeah. So it's like you can always use, though, for two senses. One is the even reading. One is the universally quantified reading. You can say, I, though, ate crabs or something. So it's like, even I ate crabs. But the other possibility is that crabs, I don't eat. That means I ate all the crabs. So in some rather neutral contexts, you have to show through intonation that one reading is preferred. And again, this is especially bad when we write it down as a sentence on paper. So you may disagree with one reading, but the other reading is also still available. But even I reading, does that mean that there were a group of people and they ate crabs and even I ate crabs as well? Yes, that means that I am probably the most likely one to be a vegetarian, like Ruth. But even I eat crabs. So again, this is kind of less likely unexpected, almost counter-exploitation. Yes, yes. That's why we build up a pragmatic scale on top of that. But that's the way of how you explain it. As far as I know, first there was this paper by Foucault Nie in 1975 on pragmatic scales. So he was talking about when you talk about even, you can say I even bought an Apple notebook. But then the hero can quickly build up a pragmatic scale. So what you're talking about is that an Apple Mac is probably the least kind of thing you expect to buy. Maybe you arrange it on the scale of pricing, which is more least likely is what I can least afford. But on other occasions, you can build up a pragmatic scale on other things. You can say, oh, he even ate crabs. So that is built up not probably on the basis of pricing, but on the basis of personal preference. So it is a kind of pragmatic scale because there is no necessary logical scale to back it. And everyone's preference can be different. But the reason you say even, then it gives the hero the reason to build it up, to accommodate the pragmatic scale, even if you don't mention it explicitly. And then the later kind of paper that I knew of was by Paul K. because he wrote a paper called Even. And that was just one of the series of works and dissertations. One recent one, dissertation is simply called Getting Even, as in English. And so in Paul K's paper, he introduced the notion of a scalar model in which he was considering all the different cases. So for Fogonier's case of pragmatic scale, it's two-dimensional. But Paul K. was thinking of cases which can be three-dimensional or four-dimensional. Although I find it hard to understand the four-dimensional scalar model. But the three-dimensional one, he gave very good illustrations of them. And then it quickly lead to pragmatic studies on scalar meaning. And that, generally speaking, you can still say it is related to scalar model. But sometimes I do find some papers or dissertations when they talk about scalar implicature. In fact, they're talking about scalar model. We're not really talking about scalar implicature because that can be, should be more rigorously specified with what they call home scales or others. Thanks. And then we will, if you have any funny issues or you wish to catch up with your friends or colleagues, please join us with I.O.E. Bar, you know. Thank you. So thank you all for your participation. Thank you for coming. Thank you. Thanks.