 there it goes okay and he's going to talk about historical linguistics well not not so much historical linguistics but the history of linguistics totally different yes actually that's what I meant okay yeah yeah but we're not going to be doing any reconstruction today but we might allude to it it's the only true linguistics history of linguistics is good too but let's see one a second I'll load let me see oh I probably don't that's probably okay that looks like a screen great oh I was actually worried okay Chan Gong can you read this are those bad colors I honestly have no idea okay yeah I have no clue okay so the presentation the presentation's title oh is that I can switch it around does that work better I don't know okay so um yeah so the talk is going to be called linguistics isn't 60 years old I'll just say there are lots of people I'm not going to name any names in this department but there are lots of people who have uttered the phrase linguistics is only 60 years old well yeah but a lot of people assume that they're not there's nothing that came before and this is partially what I want to talk about I we all know Samina's better you know and lots of people do I'm not undermining anyone who does but there's sort of I guess a general this is true of like any theoretical paradigm I guess because there's a tendency to just sort of assume because we have traveled this particular area no one's done it before so anyway let's go ahead and get into it and originally I was going to do a whole tour to force of thousands of years of linguistics that's not going to work in the time constraints so I'm going to focus mostly on Panini and grammar in classical India but also talk about some stuff since that I think are important just really things that I guess people don't touch on that much actually let me check one thing real okay yeah all right so anyway so the central findings of linguistics just to start out so language is generative it's creative we have infinite use of finite means that's the quote the Chomsky quotes from Von Humboldt language can be defined by rules we have you know linguistic rules coming into play they have an ordering to them you have basically we don't use transformational grammar anymore but there is some sense in which we have an underlying form or a formal form which has operations performed upon it or something like that and of course there are distinct subdomains everyone knows the syntax phonology morphology all of them have their own properties but they also have formal similarities etc etc and another thing that's I guess commonly touted as an accomplishment of linguistics is even though descriptively there are a lot of different constructions so we have things like passives we have things like relative clauses there's a formal if you abstract away from those there's a formal unity to all of them so you've probably heard of the you know the movement theory of everything or basically that that's the idea here like what you really have in languages is not so much constructions but different like core properties interacting in a way and we have argument asymmetries stuff like this yada yada but we had all this all before one AD and that's what we're going to talk about fact linguistics is really a really great subject because basically it comes into the the historical timeline basically fully fledged and so anyway I guess that I think I've said before we knew what exocentric compounds were before we knew the world was around then we had huge explanations and you know dialogues on them so anyway the major works of linguistics pretty much everyone knows there is this Panini in school of linguistics it's a name dropped occasionally so Panini was a Sanskrit-Gramarian in classical India you know the date is unsure I think usually 500 BC is somewhere or there about as a good date and his most popular work and probably most important is the Ashta Yayi which is literally means the eight chapters so this is like the the grammar par excellence it is literally the the most extensive grammar ever written of a language yes yeah you'll probably notice that with a lot of the Sanskrit words today but yes so yeah so it's a comprehensive grammar I mean you know there are other in the modern era we have things like sound patterns of English which is a really extensive treatment of one particular languages prosody or whatever but the Ashta Yayi is a full account of all of the phonology all of the morphology all verbal paradigms of the Sanskrit language and it also comes with a couple other sort of a dinda all of which are interesting but we're not so much going to talk about them you know he has a list of irregularly inflected words like thousands of them I think and of course a list of verb roots classified by how they're inflected and also the Shiva sutras which are interesting enough because they're literally just like 18 lines but it's a classification of basically phonemes based on how they work in the Sanskrit language it's sort of like a it's referred to in the grammar but that's you know that's another issue so after panini there was a long series of commentators on his work too I'm gonna mention in here are the Mahabhasya by Patanjali or not actually sir if that's his real name but also the Vakya Padilla of Bhattahari anyway so the guiding principle of I guess panini is what's called Laghava or economy principles usually translated and that basically guides the entire description of the language so well I think one way of putting it is a lot of people will use the example so in English we have the sound H and the sound in G and those are in like complementary distribution they never appear in the same place and I'll just say panini is the kind of guy and well their use is an example of just because they're in complementary distribution doesn't mean they're the same phoning now panini is the kind of guy who would treat both of those as the same phoning just to give you an idea of how he looks at the world and how he classifies them but the only driving constraint is simply economy how can we explain how the Sanskrit language or any other language works with as few principles as possible now the entire document the Aastad Yai is written sutras if you don't know how sutras work they're basically supposed to be as small of a expression of an idea as possible very economical I mean it's very it's sort of the opposite of a lot of Western works which are all about prolixity and sounding a really big brain the sutras about being very concise and so just keep that in mind and the panini in emo I guess or not just this isn't so much panini but you know the people followed after is first exhaustive exhaustively describe everything exhaustively describe the entire Sanskrit language every language and explain it only with economy principles only what formal rules can we use to describe this language with no other assumptions only after that you start making generalizations you start saying like how can we simplify these rules how what kind of tendencies are behind them and only after that do you make sort of theoretical implications and that that's sort of the important like the sort of data drives the theory mindset that people have I mean I'm not to say that people nowadays are the exact opposite but you know sometimes people skip the first step in here you know so Sanskrit just in case you don't know is it's the classical language of India it's one of the oldest Indo-European languages out there just so you if in case you don't know how it looks it's morphology is at three three numbers singular dual plural eight noun cases many verb tenses there are four main classes of verb inflection all of which have subclasses and moods and you know all the stuff you have a fully functioning media passive you know the phonology of course I mean even you have to think about the Sanskrit language like even to be able to describe the Sanskrit language you have to have a pretty exhaustive like theoretical view of linguistics so in the phenology you have a four-wave plosive distinction so that's like pa versus ba versus pa versus ba stuff like that for all places of articulation dentals versus retroflex you have basically just everything there's there's some things that Sanskrit doesn't have but you know oh and the other important thing about it is in the Sanskrit orthography all all the aliphonic rules are written so in English you know we have we say when you say would you you usually say would you with a jaw sound now if you were writing that in Sanskrit you would write that as a jaw in the Sanskrit language you have lots of phonological rules and they're all written and all the whole Paninian tradition is very aware that they are there so that's and in order if you ever take a class in Sanskrit basically the first year will be you just learning the basic phonological rules because they are so I mean for example one if you look at the morphology of Sanskrit one of the most common sounds to end a word with is an s now it happens that one of the phonological rules in Sanskrit is more or less you can't end a word with an s so s is become oh they become oh they become all these different things in different situations and that's in every single word of the language just give you an idea and in terms of terms of syntax we have basically for your word order ellipses all over the place and you basically pro-drop everything it depended of course that the terminology there presupposes theoretical stuff but anyway we'll get into that in a second okay so the australia again was a complete rule-based grammar basically on the whole of a language again written in sutras so the rules are written to be non-redundant so some people will complain when they're reading like Chomsky that he has too many acronyms or something like that the australia is even worse because it's basically assumed that if you if we created some theoretical construct five chapters ago you should now know what that is and we should be able to refer to them all over the place but because again the idea is to minimize how much time you're spent explaining the rules and that consistent basically having the equivalent of acronyms all over the place rules are hierarchical so if rule X applies maybe it activates rules a b and c and they can check to see if they apply and of course these are rules with respect to phonology also morphology and also the translation from semantics to syntax or something like that now in some cases I'll go into the specific specifics of this in a second you have meta rules for how rules can apply how many rules can apply what order they can apply in etc etc etc and the interesting thing that's most maddening for me is we're gonna get into the specifics later but the the australiae comes to history fully formed basically there it's very clear in fact plenty he alludes to like a long grammatical tradition before before him the different scholars he worked off of but basically the entirety of well let's put it this way so linguistics is not a discipline where all the sudden we realize that oh this new concept which no one noticed for thousands of years really basically all the core descriptive generalization some exceptions but nearly all of them come to the Sanskrit tradition just at the beginning we have no idea when people first notice them like it's not like you know someone first noticed the gravity at some period or someone first noticed you know this kind of word order all of them are just sort of there and we have they where they come from they've been lost to history because panini basically he is writing as if you know exactly what a voiced plosive is etc now anyway so Sanskrit to talk about one case study so Sanskrit case this is what a inflectional paradigm looks like in Sanskrit I've you know got rid of the duels to make it simple but of course you have singular and plural and you have eight different cases and these are the eight different cases you have in proto-indo-european nominative accusative instrumental etc etc I assume any of these no one doesn't someone doesn't know you don't need to know but they're just there and of course it's a fusional language you know so let's look at what one part of panini's theory in action that's Karaka theory so Karaka can be thought of as sort of a a semantic verbal participant maybe an argument it's not necessarily a syntactic thing I'll talk more about what it actually is in a second but it's sort of the rival of what we would think of as case theory that is it true it tries to understand why your words assigned case what cases they're assigned and keep in mind in panini's original sort of an explanation of it it's entirely descriptive he's not necessarily looking into you know what what conceptual necessity drives them but just how do they happen how do we account for them in a kind of rule-based grammar so you have basically these different Karaka categories you have a Padana which is a point of reference you have a Carmen you have a Carter etc etc and actually most of you will notice everyone's favorite Persian light verb will be very similar to a lot of these you know they're all related to the word for to do or whatever even the word Karaka but so you have things now I've actually sort of glossed these in deceptive ways to make them look familiar to you like undergo or theme don't really think of it like that I'll talk more about Carter in a second but so the idea in the panini and grammar is basically there are rules that there's a list of rules that check each possible noun in a sentence so you don't have inflections on these nouns there's some sense in which you you're presented when you want to utter a sentence or something like that you're presented with a scheme or sort of a not necessarily like it's well let's put it that actually I think I put it here no never mind we'll just go with it so the idea is there are different rules and ordered in a particular way and each word has to be assigned to a particular Karaka category in each Karaka category basically comes with a case so all nouns have to end up with case and at a theoretical level all cases have to be applied but I will talk about that in a second so if a nominal happens to belong to two of these categories let's say for example something could be a subps a substrate or a locus and an undergo at the same time what happens is that the later rule is the one that matters and I've actually ordered these in the right order so something is both a point of reference and an undergo or it's going to be treated treated as an undergo or by the grammar so these are ordered they're sort of like a hierarchy to them formally so with respect to Karaka so you know basically each Karaka class assigns a case so a Padana assigns ablative you get dative lockative accusative and the Carter the doer what does that get guesses good guess that's also the wrong answer it's instrumental nominative is actually a special case so we have a division between inherent and structural cases and in some sense that actually does exist now it punny isn't going to consider accusative he sort of thinks of that as being in an inherent case but nominative is a little different so here just sort of as a background Sanskrit sentences here number one is an active sentence so it's Devadatta cooks the rice gruel nominative Devadatta rice gruel is accusative pretty much what you would expect and you also have passive sentences and in these in passive sentences the agent is assigned instrumental and the rice gruel is assigned nominative sort of how you would expect where the instrumental is the equivalent of the byphrase in English or something like that now the other important thing for Sanchika grammarians is all of the all the nouns can be omitted so you can say Devadatta is you know well it's more like by Devadatta is cooked rice gruel you could emit rice gruel and just say is cooked by Devadatta or something like that any of these arguments can be freely alighted now the idea of now nominative case now punny he doesn't give an explicit rule for nominative case but it actually comes out you get it for free if you look at the rules more in depth the idea is the expression of meaning is non-redundant so you can't have multiple case markings or may not just case markings but marking and marking markings in the sentence that mean the same thing generally now the active verb ending which is T and the third excuse me that should be singular this is actually very interesting because in most recent versions of case marking theories like Baker who has looked at many many different totally non-European languages nominative case is a totally different animal that is not marked yes it doesn't it it it's just unmarked also it goes back to Moran's hierarchy hierarchy of cases as well so punny thought about it yeah yeah so the idea in his the way he puts it is basically the active ending there's some formal sense in that that expresses the same kind of agency that the instrumental would so by the rules of the grammar the idea is basically you can't have an instrumental and an active ending to express the same thing so basically you can't assign instrumental case if you have an active verb so what happens is the nominative comes in as the elsewhere basically you have a bear you know you have gender and number features that exist on the noun but you can apply instrumental case because the active ending is basically supposed to be in complementary distribution with it so nominative comes in as the elsewhere condition as in this was this was that happens if you can't get case in in punny's interpretation of it so punny he also has a concept now I should say there's not really a concept of a subject an object per se in indian grammar that that entire concept don't really exist but they do sort of get at the same kind of asymmetries in different ways so the original definition of the the card or the doer literally is sort of similar to our concept of agent at least how punny originally defines it but if you look at the nitty gritty it's actually understood in a much more wider way so really he thinks of Carter as something as a discourse topic or something like that in that it's not you can speak of something I mean one of the examples he uses is the pot boils now we would say that is you know some kind of undergo where that's some kind of unaccusative or something like that now in punny me what matters is we are speaking of that as being a topic we're speaking of that as being the the well the way he puts it is the where is it svetan threat Carter which is like basically the independent or the you could just say external argument or something like that to make it cruelly similar to modern linguistics but the idea is basically this sort of them at this is this isn't so much a thematic role in the sense we're familiar with as someone who does a specific action in a bird but it's more like a discourse role and this is basically how you get something like nominative case in passive sentences in his idea so the idea there would be you know in the pot boils or well think of a real passive sentence so the rice gruel is cooked by devadata so in a sentence like that basically what you would have what would happen is the rice gruel takes on the role of an undergoer and a Carter and it takes it would take instrumental case if not you know if you had this if you didn't have that the ending or whatever now anyway oh I just as an aside you could also do formally the same you could treat accusative case in the situation as a sort of as a structure or structural case in the way we treat it in that you could say basically the passive endings function like the active endings but with accusative case so if you have in this it's sort of Bertzio's generalization in that you could just have you could say passive endings or you know the lack of the active endings conditions it basically says you can't have a case alright so anyway what one of the ideas I alluded to this before is that you're not so much translating in panini and grammar from like some kind of deep structure that has all its lexical items fully formed and you're not so much transforming from that but the way Cardano puts it is you know the root punch cook for example is said to denote everything including cooking the internal conscious effort of the agent putting the pot on the stove putting the water and grains in it blowing heating etc and only once all that's done that you say Pachatti is cooking of something so the the idea here is really there's there's a wholeism behind verbal expression that is I might say Billy is cooking stew but embedded in that sentence is not just Billy in the stew but really the entire scene where that's taking place how he's doing it stuff like that and it's just we're aligning those extra arguments now what we're doing now the contrast between generative grammar and Indian grammar would basically in general generative grammar you have this these series of transformations that take you from deep structure to surface structure and what drives that partially is that verbs have lexical demands you know I you know you got to select your agent you got to select your theme something like that well in the Indian to Indian tradition basically what you have is a holistic perception of a scene an entire like you're translating directly from semantics however the brain perceives that and you're applying those different rules for case assignment based on the semantics of the scene not necessarily some kind of deep structure so the range is actually a bit wider than generative linguistics and you're starting from something that's a little less linguistic and more mentalistic and you're taking you're applying the rules basically to this idea to get a surface representation explain what do you mean okay well actually funny mention that so I actually omitted the slide so well an addendum to Carica theory so in I think it's Barter Hari he has an account of basically what we would call raising and control and stuff like this and the idea is you don't have something like equidoletion or raising or something like that the idea is multiple verbs when you have like when a word is like if you have a sentence like I want you to open the door or something like that in the Indian tradition you would be both treated as the object of want and the subject of open or whatever and basically the way Barter Hari puts it is if you ever have like two particular verbs referring to the same argument the one that is what we would call it syntactically higher or the matrix clause that always takes precedence so that would be sort of the origin of well of exceptional case marking also with it in mind that they don't as I said with like the Carter thing we might say in some sentences so like what's unambiguous raising so like I want well actually sort of even I want you to open the door in that situation I'm not wanting you in the sense I'm wanting you to open the door so but in the Panini in tradition similar to what I said with the Carter thing before the idea would be even if you're not like sort of the theme of want there's a sense in which you're referring to that like it takes on the discourse situation where it is functioning as a as a Carmen in their terms or it's not exceptional he treats it as just being an object basically it like in his idea it's not it's not really weird because in the context of language you're actually just referring to that as an object in the same way in a passive sentence you refer to you know a theme as a subject that's just how it is so it's not like they they don't really have the exact equivalence of you know thematic or theta rules the way we have them like they have similar things but they're they're it's not a problem you know for them to have something like an exceptional case market okay so related to this is the centrality of the sentence and the idea I should know every time I put a whole quote on a slide I'm like do I really want to read that out like when I actually get there but the idea here I'll just summarize it that is sort of alluded to here is like basically the idea of Indian grammarians is the sentence as a whole is the unit of meaning so we think of words as having individual meaning maybe in formal logic we talk of desk meaning desk or something like that but really without the entire sentence there's not a full interpretation now this is now if any of you guys have been to get all our good friend tgb has this lecture he gives on Phil Holm-Vunt and the rediscovery of the sentences the unit of processing and you you know if you don't if you have anything less than the sentence it's not like you know it's more difficult to interpret and stuff like this but Indian grammarians have to actually have sort of the same idea and that is the core of communication is actually the sentence as a whole and if you communicate parts of sentences those only make sense in the sense that you can recover the rest of the meaning or something like that so the communication again for Indian grammarians is not so much like a deep structure but it is we are communicating an entire scene and trying to make sense of that you know for someone else basically and you can't have half of that scene you can communicate half of that and the rest can be inferred but that is the unit of communication six minutes oh geez okay great because that I put like a failsafe in the middle of my presentation so I could quit and I think it's like a couple slides from here okay so just a couple comments on ellipsis so what is it exactly ellipsis now there are two schools of thought so of course again you're communicating whole scenes with you know when Billy is cooking the rice there's really a locus and an instrument and all these different things so what happens in basically everywhere is ellipsis so you're always getting rid of these things which are thought of as being superfluous or something like that now there are two main schools of thought for this that are sort of have cognitive theories of it one is is the parahart that I always have to like look at this word parahara parahara school of thought which is literally like a word assuming or assumption of missing words and the idea there is if you hear a sentence that doesn't communicate every coroner or something like that a caraca you have to go through assuming the individual words reconstructing the sentence literally and then you can interpret it as opposed to the assumption of missing meaning the other school of thought and these are general schools in Indian philosophy not necessarily just in in linguistics but in this idea the idea is basically meaning is reconstructed directly from a lighted output so if I say if a doctor who says to his assistant scalpel in the first school of thought in order to interpret that the assistant has to say oh he is saying could you give me the scalpel or something like that now in the second school of thought I just hear scalpel in the the situation I can reconstruct the scene and understand that now formally puny me treats the first one is being true or at least assumes it in the sense of he says stuff like you know oh where does the second-person agreement come from in Sanskrit it comes from the fact that you know there's a second-person subject in the sentence even if it's not actually there I forget the exact wording but it's basically like you can agree with things that aren't even there because he's sort of assuming that you always alighted there are a whole bunch of different constituents or whatever now one important note is there's no syntactic deletion rules in puny me ellipsis is not a process like this isn't something that's happening we're not getting rid of things in the deep structure now you do have deletion rules in morphology and phenology or whatever but syntactic deletion is either too commonplace in Sanskrit or it's just viewed as something that's not really linguistic per se so puny me doesn't address them up went off the edge that's aspect ratio for you so it's not necessarily addressed by his rules so now if puny me the thing I mentioned before is puny me has a totalistic description of the Sanskrit language but it is only necessary I mean it's only for economy principles so if he wrote something today reviewer to would probably say something like very descriptive of unclear theoretical importance and wouldn't even be able to rise it now after puny me as I said a school of commentators come out I think this is the last slide so we'll rush it all in so one of the core concepts we think of as being necessary to languages generativity potentially one of the commentators of on this actually goes into this in detail and one way or another another Sharma paraphrases it as you know how should one approach the instruction about or in understanding words should one take start by taking individual words and explain them to the totality of languages exhausted Patanjali does not approve of this technique of Pratipata Pata recitation of each word mostly because it would require several lifetimes no end in sight so instead in order to describe language you don't enumerate more themes you don't enumerate words you don't enumerate sentences you set forth a basically a rule-based grammar to generate those now the implications of this are sort of looked into by Patanjali and whatever but I guess we're out of talent time so that it oh in the part one and that's the only part you get now what is the part two part two looks like this it's the talk about maybe my shit yeah this will be part of the Luke Smith goodbye tour or something yeah anyway if you want to hang me back I'll come back okay thank you very much yeah sure sure looks like we got 45 seconds for questions so are you gonna talk I mean the rest of your stuff I want to talk about other schools like later on yeah yeah structuralist yeah well the structuralist that's that's too well-trodden people actually know about those I mean Europeans yeah Europeans oh see that that's too modern I'm gonna go I'm gonna go medieval that's so one of the things we should all be aware of this as a linguist is that there are huge language boundaries and like I mean one of the reasons realistically people don't know about the Sanskrit tradition is basically it's all in Sanskrit or the same thing is true of like medieval medieval European writings like there's so much in same stuff that people wrote about but people don't really know because like none of it's been translated out of Latin now I have gone and downloaded all these huge PDFs and like painstakingly read them because like no Latin well enough to read them and there's so much in there and I encourage everyone to like if you know another language that in which there is stuff written like read it because there's so much that's just not well-trodden in the sense that like yeah people don't even know what's there but maybe I'll talk about that next time okay thank you