Very difficult to insult people if that's what you have to do every time. I expect the Pirahua must be extremely polite. And fit to explode with inexpressible resentment.
Simon, don't be self-conscious about the hand waving. It's going to be important when you start appearing on "smart people talking about stuff" shows for the BBC.
Y'know, I was expecting the Pirahã to be mentioned before even clicking on the video... Didn't expect heccin' full sentences to be breezed through like that! Even if I'm a li'l sus of the accuracy of the pronunciation but never mind that.
@@impishDullahan There are a couple of phonotactic rules to bear in mind - /g/ is realised as [n] if it comes before /o/, and /t/ and /s/ are palatalised in certain environments, so that may be why it doesn't look entirely like the spelling. It's absolutely possible I've made some other mistakes, though! It's not something I really have any experience with.
I appreciate your video style. I get a bit shell-shocked by some of the expository videos that are frenetic, high-pitched sales jobs. The simplicity of conversation and simple word illustrations allows me to watch these videos and relax.
I agree that the simple style, girded with such a depth of knowledge is extremely appreciated. This is such a fascinating topic, but it does require a bit of thought. I think that might be why I appreciate the lack of glitz and the anti-frenetic style.
as someone who has tried taking linguistics courses in university and not enjoyed them, your videos are amazing. such great explanations in clear terms, truly has made linguistics feel so much more present in my life and easy to understand. keep it up
There is a (sign) language that was developed spontaneously, Nicagraguan Sign Language. There was a group of deaf kids in Nicaraguan in the 1970 - 1980s. Teachers tried to teach them how to read spoken spanish from lips. However the kids were unable to learn the language. During the breaks the kids were playing outside with themselves. They started to communicate with themselves by hand signs. Those signs evolved into a full blown sign language with it's own grammar.
The thing about Pirahã having no basic color terms is not as weird as it sounds. They do have color words; they're just the names of plants with those colors, similar to how we have "orange", "violet", and "lilac" as colors. The thing about them not having numbers seems to be pretty much what it sounds like, though. They aren't the only culture that doesn't count, but most languages like this still have a few small numbers. That being said, Pirahã does have a word for "one"/"a few" and another for a larger number. It's probably not that similar, but it does remind me of how a German word for "some"/"a few" is "einige", which I think is related to "ein(s)", the German word for one (which is cognate with English). I've always wondered how that change happened since I noticed the connection.
It has also been shown to some extent that practitioners of languages without numbers have immense difficulty in understanding abstract thoughts. Im sure there is some debate on how related this may or may not be to the nesting of sentences.
@@Sugmatron The problem with that is that they don't necessarily have trouble with numbers BECAUSE their language lacks numbers. An equally valid conclusion is that they have difficulty with numbers because they didn't learn about them as a child. That is, the fact that the language lacks numbers and the fact that the people have problems with numbers both stem from a third factor: The culture doesn't count. This is a classic case of "corrolarion does not prove causation". FYI, the above is an argument against what you said being evidence for the Sapir-Worf hypothesis. I know you didn't explicitly say it was, but it seemed like you were implying it. Also, I notice you actually said "abstract thought" and also "practitioners of languages without numbers". I don't know what other types of abstract thought you were talking about, and I don't know what other languages that Pirahã you're talking about. I know that the Pirahã specifically supposedly don't care much about things that far away from them in space or time, e.g., ancient history, legends other countries, etc., unless it specifically affects somepne they know or someone that someone they know knows. The only problem is, I think a lot of native English and Spanish speakers kind of like that, so I'm not sure how Pirahã-specific that really is. I'm also just going off of what Everett says for a lot of this.
@@Mr.Nichan I agree in that correlation is not causation. But I'm not certain that one can easily separate culture and language in such a way where you can decide if it was the "chicken or the egg" so to speak. I was not speaking directly on that hypothesis, although related in hindsight. Just the idea that the linguistic (or cultural, as much as it could be separated or attributed) lack of counting and other forms of abstractions may have some relation to the lack of nesting sentences. I said abstract thought as there was a study years ago where researchers worked with established language practitioners of languages without numbers along with non-standard sign language practitioners in Brazil (I'm not exactly confident in the nation, I would have to find the study once more). The users of these home-made sign languages which were deaf from birth and had no access to a modern sign language, along with the standardized spoken language practitioners showed an inherent lack in understanding of abstraction. I specifically remember a focus of the study being how the practitioners of the sign languages could use money and make change as easily as the average person, but once numbers, or even time was abstracted away from physical objects or actions the understanding went with it. Quite interesting stuff.
@@Sugmatron Those particular abstract thinking skills you mentioned were pretty quantitative. I'm not sure if there's any reason to think that difficulty would translate into non-quantitative types of abstract thinking. I also wonder how they were attempting to abstract it away from physical objects without switching languages. I suppose they used numerals and something to do with counting time?
There is so much of interest to learn from these videos but just for starters lets all try to display the humility shown here. Whatever we do , lets ne a bit more like simon.
I’m from Hawaii, so it was cool to hear you mention Hawaiian creole (though we just simply call it “pidgin” here even if it’s not actually a pidgin language). It’s pretty interesting because growing up I never thought of it as a different language. It’s often looked down upon as a lesser form of English, or broken English, and people who speak pidgin primarily (I don’t, but many of my older relatives do) have similar problems to the ones mentioned in your AAVE videos. Code switching is very common, and even among people who speak primarily pidgin there’s the mindset that it’s lesser, or uneducated, or low-class-myself and many of the people in my age group (at least the ones I know) speak standard English primarily, and it’s not uncommon for my older relatives who speak pidgin to say that we “talk good” and often older people who speak pidgin try to encourage their kids to “talk good” because they think it will help their kids with school and jobs and stuff. I admit that growing up part of the reason I preferred standard English was because it seemed more proper and educated (plus other factors, of course, like it being the primary language of education, the fact part of my family is from the mainland, and the fact that most of the media I consumed growing up was from the mainland). As a result I still don’t feel that comfortable speaking pidgin, because it feels like I’m putting on an act because I never actually spoke that way growing up, but after taking some linguistics in college and learning more about the language and it’s rules I found a new appreciation and perspective on it.
Cheers to whoever sent Simon the mic. I must admit, I thought about doing this when the phone came into play, but I thought it might be a bit presumptive. Well done whoever actually had the motivation to do it!
Reminds me of my favourite sentence, which is an example of preposition stranding: "What did you bring the book I don't want to be read to out of up for?"
Chomsky's whole point is to demonstrate that language is more a biological phenomenon than a cultural one, while Everrett would be the reverse. I did a bunch of Chomskyan linguistics in undergrad, and I'm pretty sympathetic to Chomsky's ideas. I think his starting point makes sense. The marked cognitive difference between humans and other primates (including humans' ability to use language) must be due to a relatively recent evolutionary event. If that event produced the biocognitive capacity for language in humans, then surely that capacity *looks* like something, right? It must be, in some way, tangible. The capacity for language that arose is ultimately biological, so it has limits, patterns, and so on, that are hardwired into the human brain, just like how any other biological system in the body and brain also has limits and patterns. And since that innate hardwired capacity governs how language arises and develops and is used throughout human history and across the world, we should expect it to exert the same constraints on all human languages. So it all sort of makes sense to me. In saying that, since undergrad I've since become more convinced that there is a lot to be said for Everrett's camp, which sees language more as a cultural tool. There is lots of good evidence that culture drives language and that language reflects culture, which I won't go into here. Though, this cultural tool take still has to be stacked on top of some biocognitive capacity for language. I guess it's possible, as you mention, that this biocognitive capacity is just a combination of higher-order pattern recognition, symbolic reasoning, and other cognitive abilities. And the point that children may quickly pick up and make sense of grammar based on word distributions, not anything innate, is a good argument here. Thanks for making me think. You might also be interested in exemplar theory, which says that speakers (including children) store words that they hear in the mind with all their acoustic, social, syntactic, and other contextual information. So when people (again, including children) speak, they don't piece together grammatical sentences (and develop grammaticality intuitions) based on anything innate, but rather on the wealth of exemplar words and sentences that they've amassed.
I don't think it's completely fair to divorce the phenomenon of language entirely away from its biological substrate. However, Chomsky's single-step hypothesis is far-fetched, in my humble opinion. While a gain of function mutation to the coding region of the ancestral variant FOXP2 could have certainly accelerated the development of grammar in ancient humans, mutations to genomic regions that regulate how FOXP2 is used by neurons could have an equally if not more potent effect on the development of neural circuitry for speech. In fact, the two mutations in human FOXP2 previously thought to be responsible for human verbal skills are shared by Neanderthals. However, there is a mutation in a regulatory region of the gene that is present in humans but not shared by Neanderthals and Denisovans. I think it's more likely that several moderately advantageous mutations had accrued in the ancient hominin lineage that could confer language, but humans have managed to acquire (at least) one more that made learning and systematising language more efficient. However, either hypothesis is incredibly difficult to prove.
It does seem like the trail goes cold once you try to actually describe the biological basis of a universal grammar. Exactly which proteins and genes are responsible for language? Chomsky's UG is, at the end of the day, based on a biological assumption. Most of his work is not about biology; it only uses the biological assumption as a foundation. From there, Chomskyan linguists launch into the way in which grammatical sentences must be generated in the mind if there is some underlying UG.
@@nickhight I agree. Chomsky is just using biology as a premise for his argument. But that didn't stop a bunch of biologists from trying to test those assumptions though! I think it's an exciting field.
@@nickhight While that is true in a way, if you look at Syntactic Structures, Chomsky's first book, you see that what he is attempting is a mathematical rather than a biological approach. You have to remember that he was reacting against the Behaviourism that was widespread at the time, so I see his theory as a specialized theory of behaviour, and to that extent divorced from genetics and mainstream biology. The point of contact would be neurophysiology and neuroanatomy, really, rather than genetics.
Does it have to be an evolutionary event though? Couldn't you just argue that the social nature of humans (and other primates) the strong ability to teach can basically "snowball" the development of a common tool such as language? I think you can apply this same line of thinking to prehistoric tools and even see it in action today with new tools such as smartphones and other technology, there is a very strong ability to pass knowledge and understanding down, and this dissemination of knowledge fundamentally makes primates prone to adapting their behaviour in ways that exceed the speed of genetic evolution because the person learning from another doesn't need to put in the same effort as the person who figured it out in the first place.
Very strong evidence for innate universal grammar is provided by the case of Nicaraguan sign language. During Nicaragua's long civil war the education system for deaf children collapsed, so a generation of them were just brought up isolated at home, and also the teachers had retired or were dispersed. So when peace was restored they decided to provide education for deaf children, and collected a number of them together. However, they needed to train the teachers, and decide what system they were going to use. So at first they left the children alone while they discussed these matters. Amazingly, these young children, who didn't know each other, invented their own language, a signing language. This was so successful that it has now been instituted as the official deaf sign language for Nicaragua. Analysis shows that as with other signing languages, the grammatical structures are the conventional ones you might find in a spoken language. The idea that this was generated by culture seems hard to justify or accept. Congratulations on the channel, by the way.
Michel Dansel, under the pen name Michel Thaler, hated verbs so much he wrote a whole novel without using any. He still manages to express his meaning quite comfortably. But I suppose the point is that he had to work really hard in a language that already prefers nouns to verbs to contrive a way of doing so. en.wikipedia.org/wiki/Le_Train_de_Nulle_Part
You're very modest. I personally believe in UG - but I do think that the evidence for its existence is theoretical. That is, it is a logical construct. If there were not a language acquisition device in a child's mind, we would expect errors that do not occur. From that I do think that the binary branching model is the best we have for how language can plausibly be processed in the brain. Even if our brains operate just like Google translate in every respect (for instance), then surely that is "universal grammar" as well. It would be an overhaul of the Chomskian binary branching model, but it would still be part of the biological endowment. Universal grammar is the scientific name for whatever it is that helps us acquire language, and is specific to human language (rather than an alien one). I hope this makes sense and that I haven't presumed anything incorrectly.
I always loved Chomsky's activism, but universal grammar has always appeared to me an unbelievably bold a claim given the available evidence. Science proceeds because of the certainty granted to it through the poverty of it's epistemological means, and not by the explanatory attractiveness of it's hypotheses alone.
I feel pretty comfortable not developing even a tentative opinion on this question, partly because for some of the points you described I can hardly imagine what the difference would actually look like.
Just one very small nitpick. "Who did Jack ask when Jill saw?" can, as wildly unnatural a construction as it is, be grammatical. The verb "to see" is technically ditransitive. So, we could say this if, for example, Jill had been blind most of her life and saw for the first time. Here's an example of what you're referring to: "Who did Jack ask when Jill inflicted?" This sentence should immediately provoke a "what?" response because you can't rationalize it no matter how hard you try. Something needs to be inflicted, whereas one can simply see without necessarily needing an object. I'm hoping that the response to this comment, if it gets one, is an "I see." ;)
I very much appreciate the slides! If there's an auditory version of dyslexia, I think I have it. Reading along helps me so much! It's also nice to see your face, too, so it feels like a conversation not a lecture. And the nature shots are beautiful and peaceful. Yay for balance!
My university is very Chomskyan, so I am very biased, but I find the fact that damage to certain parts of the brain (Broca's and Wernicke's areas, to be precise) leads to aphasia, certain parts of language being damaged (language production in the case of Broca's aphasia and language comprehension in the case of Wernicke's aphasia) a strong argument in favour of language-specific areas in the brain.
I remember reading an interview with Chomsky in Omni magazine about forty years ago. The interviewer kept asking him how were were going to communicate with space aliens when we make contact. Chomsky quite reasonably said he had no idea because he didn't know what the aliens would be like. Chomsky also casually mentioned the term "universal grammar" as if it were assumed the interviewer was familiar with it--which led to another round of questions about aliens. Chomsky corrected him, saying universal grammar was only universal to humans. The interviewer replied, "Maybe we shouldn't call it universal, then."
I have a layperson‘s fascination with linguistics (and specifically Old English), and you make it so much easier to understand than other TH-camrs. The videos only get better as time goes on. Can’t wait to see what the new stuff looks like.
Just in case anyone else is learning German like I am, remember that the 9:05 sentence is very literally translated and therefore doesn't really work. It would have to be 'ich zerbreche...', 'ich zerschlage', 'ich zerstoere', 'ich mache das Fenster kaputt' and so on. The zer- is necessary and important as it's a verbal prefix that suggests some sort of dissolvement or destruction. Very interesting video though! Keep them coming!
Nice video, Simon. In this kind of discussion, it's important to keep clear in our minds two distinct issues: whether humans have an innate capacity for language; and whether there is a universal syntax. Some of the commentators in this board seem to be confusing the two questions. I find Everett's argument convincing, namely that there is no universal syntax. Children do seem to have an innate ability to quickly deduce linguistic patterns (rules) associating sounds with concepts and combining words to make sentences. But the rules themselves seem almost completely arbitrary (i.e., cultural); except to the extent that they must accommodate universal human cognition.
I definitely lean much more towards the Chomskian side. So I disagree with you on that. I'm sorry, I hope you don't hate me for it. What convinces me the most about it is the fact that language as a thing is so universal. Like, the Pirahã language may be very unusual, but it is still a language. Where a population is hearing, there will be an oral language where sounds are made with their mouths to represent individual semantic concepts that are later placed linearly next to one another in extremely complex and specific ways to encode meaning. And the same will happen with a sign language if the community is non-hearing. When you say that linguistic universals may be a product of other cognitive processes that just naturally coalesce into a specific way of using language, I see where you're coming from, but the difference is in just how incredibly quick and unconscious the acquisition of language is. Other cultural phenomena require a certain degree of active learning: a child will not learn to bike, to read, to sew or to play the piano if there isn't someone who is telling them what to do and how to do it at the early stages, even if they get to master that activity and do it like it's second nature, they had to observe diligently at the onset of learning. Some kids may do some of these things without initial guidance, but not all of them. Language, on the other hand, is clearly engaging some of those cognitive processes, that may be more general, in a very specific, goal-oriented way, to force a child to acquire language simply from exposure, and the stages of acquisition are surprisingly consistent. If our brains were just good at certain cognitive processes, and we use those things to acquire language, without a specific module that specifically aims those processes to the acquisition of language, I feel like we should see MORE variation. I know that few things are truly universal, and language is clearly formed from outside stimulus that is culturally determined, but even if some things aren't universal, I just couldn't convince myself that they are as "obvious" as to appear in so many languages that have nothing to do with one another, and haven't been in contact for millennia. I encourage you to look into conlangs, which are constructed languages, and to see the incredibly different ways people have come up to encode meaning, that you just intuitively know couldn't be possible in a natural language. I especially recommend Ikthuil, Lojban, Kēlen, Fith and Rikchik. When I think of these conlangs and compare them to natural languages, I can't help but wonder why natural languages just don't evolve to allow more levels of complexity, like Ikthuil, or why they don't organize predicates like Lojban, etc. And I think it's partly because we don't "just use" what our brain has to offer when we acquire language, we use what it has to offer in a particular way, which encourages and allows certain things and disallows others; otherwise you'd see more diversity. But that's just my opinion. Perhaps those "obvious" ways in which to use our cognitive abilities really are obvious, and that's why we tend towards them, and maybe our languages really are the most effective ways of imparting information, and they have these things in common simply because they're the most direct way to do it with the tools we're given. I'm just skeptical of that.
I think the biggest "proof" for chomskyism is how sign languages arise naturally, even among illiterate people-i.e people who have theoretically no exposure to language, manage to use the tools at their disposal to create meaning, to translate there thoughts into expression.
Ah, a simple short disagreement, fair enough *Clicks read more* OMFGGGGGGGGG Jokes aside, great argument. I'm gonna play devil's advocate to both arguments though cos I don't know enough to make points of my own lol. When Simon said part of his reasoning for not siding with Chomsky too much was because a language 'module' doesn't appear to be strictly necessary, my physicist mind was delighted because that's how I like to think of things whenever possible. However, I don't think that argument is particularly relevant here because evolution isn't a perfect science, there are countless examples of over-evolution and 'dead-ends' etc and with how important and common language is to humans its not a stretch to think we might have one. Despite that, I do see a flaw in you suggesting that there is no, or little, 'active learning' and 'diligent observation' in regards to language, which I have inferred to mean conscious, logical, observation-based thought. From what I've heard we start 'observing' language from in the womb and are clearly constantly exposed to it afterward. I think its fair to say that counts as extremely extensive observation. As far as I understand, our brains, especially when so young, are extremely malleable to stimulus. While having a brain region for this this function seems to make sense as an argument why this happens earlier with language than your other examples, if you think about what stimuli are present to a baby/toddler speech is one of the biggest and so it follows that the baby will spend it's time and thought on puzzling out those stimuli first. This hinges on whether you deem a baby capable of conscious thought, I would argue yes, in a different way to adult thought but still concious nonetheless, but I'm not a neuroscientist and don't have evidence to back that up for or against. So I know this isn't a flawless argument by a long shot but I thought it was worth mentioning to provide some counter to what you were saying. Kind of ignoring all of that and going back to basics though, I thought neuroscience had already decided we do have areas of the brain for language? I assume the Chomsky/Everett discussion must be about the specifics of their function then and how developed they are? Like whether it extends to things such as a natural distinction between nouns and verbs talked about in the video or am I completely wrong?
I’m not a scientist by any kind, just a passionate by linguistics, but I want to point few things: children until few years don’t learn the language in a way that an adult does it - their brain learn by mimic, while an adult brain lost partialy this ability; that’s why is so easy to learn new languages at early age (I have such an example in my child, comparing to myself); I make a very clear distinction between conscious speech and simply speech - in my humble opinion most of the people have a simply speech, making a lot of mistakes and having a limited vocabulary; I truly believe we are living computers, but even so I always wonder how grammar appeared! It is a concept so complex that I doubt it is human! 😁 Simon himself seems to be from onother planet 😄 joke aside, I admire him and I enjoy every video he posted ❤️
I love your nature's shots! Everytime it's so spontaneous and mystical ! :P It's a good topic ... I don't know if languages is human based, i think we definitely have a place in our brain ''Made or evolved to be'' for ''verbal languages''. But if only a dog or a cat could have a human's mouth and vocal cord.. I would be sooo curious to see it talking because i have cats and sometime they respond and talk to us like we do. It's mind-blowing! A pretty good question would be ; How well do you think your dogs/cats understand you?? I think we'll be shocked! And it could maybe answer some of the languages origin's questions Sorry for my grammaticals faults , i am a Canadian French... I should not say that, i'm gonna be beaten down xD
As I have grown older, I find that I'm a frustrated linguist. As a wee lad playing in my sand pile I wondered about language. I may have just heard the story of the Tower of Babel. I was interested, even at that young age, where it began, what it sounded like and how did it become my vernacular. I wish I had realized I had this interest when I could have done something about it. I really appreciate your videos.
i havent watched your videos in months (no particular reason hahah, ive just been busy and i like to listen to music or more mindless stuff like livestream VODs while i work) and these "how did language start" ones are my first few back--i cant believe id forgotten how calming and informative and wonderful i find your videos!! thank you for making this series, i do lots of worldbuilding for fun and your channel is a great resource for my art c:
Really fantastic stuff. The best overview of the subject for newbies and intermediates in this realm. Glad to see you read Don’t Sleep... I would be interested to see a video just in that book.
I like the slide heavy vids because it aids with digesting the information but I am also a fan so feel you have to express your work in a why and style that represents your vision for the channel
When I started to study linguistics at university, I expected to be told of systems of communication that did not conform to universal grammar in contrast to human languages. After all, a good way to understand a concept is to consider its opposite. I didn't hear of any examples. I can think of some possibilities myself; mathematical notation, computer languages and, perhaps a good one, bidding systems in bridge.
What exactly do you mean by "mathematical notation"? If you mean logical notation that is used to express concepts in mathematics (like S-PL for example) : those are basically artificially constructed langages, with very strict grammar. And with regards to "computer languages" do you mean programming languages? Because those fit into a lot of definitions for universal grammar too.
@@shiniri2647 By mathematical notation I mean the ordinary symbols used to express formulas and equations. By computer languages I mean programming languages which have a syntax very different from natural languages. I have used many of them, I am a retired programmer, and found they are not like human language. The only possible exception is COBOL whose designers intended it to be written and read in something like English. This idea was quickly found not to work because programmers complained it became too verbose so that simpler non-English like structures were introduced.
@@thomaskember4628 I'm in no way qualified to comment on this, since I've never studied computer linguistics, only regular linguistics, maths and a couple of programming languages, so what I'm saying is based entirely on what I made up myself, and cross-application of my knowledge of those fields. But I have found that the programming languages I know have similar syntax and semantics to mathematical logic (meaning the classical first-order) which in turn fullfills the definition of a universal language by Chomsky, namely that there has to exist a finite set of rules for the organisation of the language. Where exactly do you think the difference lies then ?
With Daniel Everett you mention something that originally started my interest to go deeper into linguistics. Just funny to have made this circle for me ;) And I just recognized a very interesting linguistic feature of your videos (and probably other videos on youtube): The very interesting mixture of the poles of the conceptual continuum of orality and literacy of the language of your videos. What really triggered me here was the train, when the rather scripted section was interrupted by "wait, there is a train coming". Former colleagues researched on what effect that might have on students when a teacher uses language more on the side of orality of the conceptual continuum or on the side of literacy. The quick answer: It is not that simple! :)
Language seems to be derived motorically from a more general movement feature, the „Willkürbewegung“ studied by the ethologists. It’s functional polymorphism. The evolutionary trends converge to language independently in parrots, corvids, toothed whales and primates. But it’s more sensible to view at language from a teleological perspective at the beginning. Language is mainly a feature that allows to instruct a conspecific of a precise to do. Honey bees are a model organism for a simple language. Their waggle dance language delivers a decoupling from spatial and temporal constraints imposed by pheromonal or tactile information „techniques“. When a corvid raised by hand tells you „Stay away!“ it‘s an imperative approach . When it says „ I‘m hungry“ it‘s an declarative one. All appetitive behavior is structurally more complicated than avoiding. I personally think, that language is a „child“ of the optical sense, because this form of perception is not restricted by short informational events like acoustic, tactile or olfactory. The „mild omnipresence of light“ (Novalis) opens up a categorical space, that‘s nearly boundless. The german term „Bedeutung“ encompasses such a new category. „bedeutungs“- sensitiveness has been studied by Auguste von Bayern in jackdaws. These corvids are able to reason about the intents of their conspecifics by following their line of gaze. This is one category of indirectness, that‘s only possible under the conditions of optics. Others are „option“, „expectation“ and „objectivity“. The nervous system therefore is solving differential equations, like Sabine Hossenfelder pointed out in a talk on TH-cam. I think it makes sense to combine linguistics, reasoning about the capabilities of an optical sense and Category Theory for a deeper understanding of language. By the way, the decision between verb and substantive is not a linguistic trait, but a universal one between type and function. „We routinely use types to express our high level intent of our programs“ (Nadia Polikarpova in „Type-Driven Program Synthesis“).
A huge number of linguists would argue for a genetically-encoded, universal grammar. A huge number of linguists are wrong. They are the same linguists.
If I could be allowed to show off one more time; the sentence: "Where John had had Mary had had had, had had had had the examiners' approval," is perfectly grammatical and semantically straightforward in English.
But that's cheating a bit isn't it. If "had" is plums and "had had" is prunes, prunes pluperfectly got the nod. What's impressive is that most people are able to see through the trickery and understand it straight away, especially if it's spoken, which I suppose is testament also to our expressive vocal ability.
This is my first time seeing this sentence, and as a native English speaker I'm not finding it very straightforward at all. Could you expand and explain what the meaning is supposed to be?
@@baronOdaighre Sure. Two people, John and Mary, have completed an English exam. In one sentence John had written, "Sid had some bananas", whereas Mary had written "Sid had had some bananas", which the examiners preferred. Therefore where John had had Mary had had had, had had had had the examiners' approval,
It's interesting that you brought this up, Simon. I've been teaching myself Ukrainian and sometimes I will have to choose the missing word in a lesson and I don't know any of those words. First, I see if the word is probably a verb or a noun, for example, and then I look at words I do know and see if the new words are spelled similarly, like maybe it's the plural form of a word I know or something related to something I already know. Sometimes I just shot in the dark guess. 🤷♀️But it's interesting how all languages share some very basic rules with each other and even if you don't know the language, you still have some basic knowledge of sentence structure or that verbs have different forms.
Very nice video, as usual. For those interested in the subject, I'd recommend two older pieces (in English) that, without exactly arguing for or against universal grammar, offer insightful critical remarks on the frameworks from which the topic has been debated. These are Eugenio Coseriu's essay "Universal (and other) universals" (available at www.romling.uni-tuebingen.de/coseriu/publi/coseriu128.pdf) and Ernst Pulgram's review of Chomsky's 'Language and mind.'
We think recursively because every verb and noun in our inner dialogue comes loaded with associations, memories and emotions; in order to form a spoken sentence we must submerge all but that which serves the intent of the moment. All else is, as it were, put in parentheses of silence.
I’m no expert, but the “arising from human thought” perspective sounds like a lot of things I’ve heard ascribed to Chomsky, eg “language is just human thinking translated into sound”, and “human thought is symbolic and grammatical”. Am I missing a part of the argument here? Really cool video and interesting topic btw.
I think you may have a point in that there is a school of thought which says that human "rational" thought is based on human language. The "arising from human thought" argument that he is discussing in this video, on the other hand, is more of the idea that human language is just one application of human thought, which is not fundamentally linguistic. The direction of the causality is different. He only briefly touched on the viewpoint you mention when he discussed that some people seem to "think in language" constantly, some apparently don't really ever, and most people are somewhere in-between.
Given that we're born able to mimic the rhythm and cadence of our mother's tongue (our 'accent') it seems we have a universal 'music' module, but acquire our grammar as we do our phonemes: by fitting them into the tune by trial and error. news.bbc.co.uk/2/hi/health/8346058.stm
Isn't there a lot of neurological evidence for universal grammar, namely TBI in different regions or strokes in different regions either causing severe language impairment or no language impairment?
Have you read John Truscott’s Consciousness and 2nd Language Learning? He makes an interesting argument for the dedicated language learning module idea. Or if we want to consider rational processes vs. UG, then what can we make of English attributive adjective ordering? The general “rule,” as I understand it, is that adjectives more intrinsic to the noun go closer to it: “big green tree,” since trees come in various sizes but are generally (culturally?) perceived to be green. If we are presented with “green big tree,” our brains inevitably scramble to find a context in which this makes sense, such as some bizarre horticultural experiment with a row of small varicolored trees and a row of large ones: Since the speaker would need to distinguish which row he/she was referencing, you might get, “But what about the green big tree? It’s doing nicely, don’t you think?” Thank you for your videos. I greatly enjoy your subject matter and the clarity you bring to it.
Based on how evolution works, I think it's far more likely that language evolved out of a collection of programmes in the brain and not from a singular, discrete programme. Much of the architecture of the brain is deeply conserved across vertebrates. This is clear not just from comparative neuroanatomy, but the central nervous system across vertebrates have similar patterns of development from embryonic tissues. Nature tinkers with extant gene expression programmes to build more complex emergent properties if it is so advantageous. Moreover, neurons have special evolutionary pressures on them, and thus have additional mechanisms to generate diversity in gene expression patterns to make such elaborate functions robust (e.g. alternative splicing, stop codon read through, 5' UTR regulation, etc.) Edit: Clarifying that this is mostly my perspective as a person who studies genomics and the evolution of neural circuits (I do not have expertise in evolution of language or neurolinguistics per se).
Re early childhood learning: young children can and have learned a wide range of astonishing skills, when their environment has been rich in that particular practice. Musical instruments, painting or drawing, other craft and technical skills which are distinct in many ways from spoken language. Is there a piano machine in the brain? An African drumming machine? A pen and ink illustration machine? No there is not. There are a set of resources and structures in our brain and body which have historically accomplished an amazing array of techniques and crafts, and can clearly be harnessed very early, shaped almost entirely by what matters to the humans around the person, and what they have to offer materially. I just don't see how fast achievement in language (also not universal of course, though nearly so) is evidence of a structure uniquely human and uniquely for language. Also perhaps worth questioning whether Chomsky's idea A: lives in a larger project of making sciences "harder" to increase their status, and B: has an unfortunate association with some nasty ideas about disability and neurological difference, which have emerged from the great man's mouth on more than one occasion, and been backed by his followers. I have twice had dime store Chomskyianism thrown at me as an argument against supporting communication strategies for non verbal adults, so that's fun. The point is that a sociolinguistic view of language propogation is not only probably more correct, it is also more humanistic and fun and full of life, less reductive, more multiplicative.
For whatever it's worth to you, I think your face is a perfectly nice thing to have on screen to accompany your mellifluous voice. No nature scene can compare. ◡̈
3:25 Technically this sentence could make grammatical sense, if say, Jill had been blind but was suddenly able to see again, and Jack had gone to ask someone about it. Of course, this completely changes the meaning of the sentence.
Just to make the point, Chomsky himself clearly distinguishes what he meant by the term (which he concedes was poorly named) "universal grammar" - essentially the fundamental observation you cover at the beginning of the video - and the field of divergent theories that spawned from this base observation. In his book "What Kind of Creatures Are We?" he goes into a lot of detail about this, and also notes that he considers a few of these theories more plausible than the others but isn't "married" to any of them and has discarded a fair amount of his own work in the field based on new evidence (generally how scientific inquiry is meant to work), but that the basic observation stands. So it always strikes me as odd when people argue that "Chomsky is wrong on universal grammar" when what he meant by it is so different from, far narrower than, and comparatively uncontroversial to what is actually argued against, and what is argued against is itself usually a strawman version of the multiple theories in that field.
On the evolution of language scale, I agree those are somewhat extreme views. I think I lean a little more to the Chomsky side. I don't think there is a single language handling brain module, because there are at least a couple (maybe three or four, I don't remember exactly) different areas of the brain that handle language related functions, but also the portion of the brain that handles the articulation of the mouth, tongue and larynx must particularly developed to allow for language, which blows the other guy out of the water. I don't know how long ago the Chomsky position was formulated. It would have been reasonable in the 70's, but we've learned so much about the brain in just the last twenty years that the position isn't tenable. I would be curious to know if he has a different position now. Mostly because he says a lot of clever sounding things that are quite daft, and I wonder if that extends to his actual area of expertise.
Can you explain more about the gaps in the argument that the lateralisation of the brain led to human language? I haven't read much about this topic since Julian Jaynes wrote about it a while back.
@@JonnaaM Hola! me imagino que tiene que haber. Yo te diria que busques bibliografia sobre linguistica de las universidades - son gratis por lo general.
I totally love your presentations, Simon. Although I certainly support your interest and trajectory in archeology, having an abiding interest in Greco-Roman Antiquity, I also think you’re a natural for linguistics or consciousness and theory of mind studies, which appear more and more to be inextricable from the evolution and study of language.
I have a gut feeling that the study of machine learning, in particular language models, will help solve this mystery. Because neural networks do bear similarities to real brains, and in some cases are essentially blank slates full of neurons, inputs, outputs, hidden layers, working in conjunction with other neural networks, and they do acquire emergent properties, in some cases literally evolve them. Similarly, existing language models, such as GPT-3 can become capable of other types of abstraction, for instance being surprisingly good at arithmetic.
However, the assumptions that underlie the language models are not neutral. So, rather than solving a mystery, the result might be sophisticated confirmation bias.
People have been proposing the same thing for about fifty years and there doesn't seem to be much progress. The machines get better and better at imitating written language in particular, but it's a bit like the assumption you can train apes to talk using sign language. Looks like a good idea, but when you try it it doesn't work.
It's an interesting prospect, I lean towards the confirmation bias theory. Because although we can jokingly say some languages can feel like maths, Latin springs to mind, they're generally not so mechanical in nature.
@@christopherlord3441 the one difference (use of AI versus trying to impart human communicative capacity to an ape) is that you can attempt to approximate the cultural impact over time of developing and passing on novel methods of communication that then overt time get reinforced , adjusted, or abandoned. Not that AI might not be a dead end, but I wouldn't rule out it's potential future role in testing hypotheses of language development. It reminds me of the shift in computational in brute force chess analytics like IBM Blue, vs the AI generationally self learningethods of AlphaZero. I'm hopeful sound scientifically rigorous use of AI can sid in adding insights in this field.
@@user-eq8ww1gr6v Yes I expect so but analysis of neurophysiology and neuroanatomy plus modelling of real human language will get there first in my opinion. You might be right.
I've never seen any evidence that animals don't have language. Or that they don't use sounds to communicate. One time I was walking to work and a baby bird was on the ground, meanwhile the whole flock of related birds were freaking out about this baby bird. If you tried to get near it they would dive bomb you. They were chattering and chirping like it was the biggest tragic story since the Hindenburg.. I'm sure one of the things those birds said would translate into english as "Oh the Birdity!!"
The question would be whether or not the sounds are reproducible without the external stimuli. That is to say, when a person laughs, there can only be on meaning to that sound: it has an intrinsic value which cannot be "faked" in the sense of it always denoting a particular emotional state. Human language however assigns sounds abstractly, an often arbitrarily: the sounds cannot be relied upon to have the same meaning regardless of circumstance, and thus requires speakers to know which sounds are being applied to which concepts. A human scream, laugh, shout, etc. are not arbitrary, and have intrinsic meanings, which is what most other animals' vocalizations achieve as well.
@@mr.flibblessumeriantransla5417 Vocalizations aren't the only way to communicate. There is sign language for instance among us humans. Furthermore, for instance among dolphins they are able to communicate sound pictures to each other. For instance, say a dolphin uses his echo location to detect an odd object like a crash plane in the ocean. It can reproduce the auditory clicks that correspond to the sonar picture of the crashed plane. Essentially they use sound to make pictures they process visually and they can communicate those echo signatures to other dolphins essentially forming a picture of the object in the minds of others. That's communication. That's language.
@@malrec I didn't mean that there were no instances of other animals using what could be classified as language, just that its very uncommon among most species to produce vocalizations that aren't merely representing emotional states. Dolphins, as you mention, are a great example of another species which creates new sounds and assigns them specific meanings, hence: language. As for your OP, it would depend on the species of bird. Most wouldn't have the cognitive capabilities to formulate sounds and assign them abstract meanings, though some of the more intelligent species likely could. It's also worth noting that "language" and "communication" aren't necessarily synonymous. Language relies on the ability to convey information within a predetermined framework, which requires both the speaker and listener to agree upon the chosen sounds. This is different from simple vocalizations which express emotional states, like crying, shouting, laughing, screaming. While there's no precise "line" dividing "language" from other vocalizations, the point is language uses sounds that are inherently _unreliable_ in that if the recipient doesn't also agree upon the application, then no communication can occur. This means that language development likely goes hand-in-hand with cooperation, with greater sociality creating better opportunity for ever more intricate meaning to be expressed.
My dog has different bark sounds for "someone that doesn't belong there is outside the house", "there is definitely something dangerous outside the house", "uncle has come home, I don't like him but unfortunately he lives next door so he may pass" and "dad has come home and is outside the house" followed by "please open the door for me, so I can greet him and make a fuss"... This is just the translation in English, because she barks in Italian. By the way, I'm German and I discovered so many things that are impossible to express in Italian. It feels as if they lack half of the words I would like to use. But, it might probably be the same the other way round. I find it fascinating how language forms our thinking and causes or expresses different mentalities.
Music; singing developed into language and yes the brain cells linked to sounds and notes are older, evolutionary speaking, than the language center cells. Plus we've found bone flutes that predate when language was thought to have developed.
11:50 This reminds me of stack-based programming languages (which includes most assemblers) or reverse polish notation, where instead of (1+2)*(3+4) you'd write 1 2 + 3 4 + *. This eliminates the need for parentheses as the order of operations is unambiguous.
I've been reading Peggy Mohan's Wanderers, Kings, Merchants: The Story of India through its Languages, Penguin / Viking, 2021. Although Mohan writes in a way that's accessible to non-linguists like me, I still paused midway to look at your video as it covers some similar ground, in order to understand the subject better. If you do have the time then please review the book I mentioned. I, at least, would certainly read it very avidly.
Another thing about "universal grammar": Being encouraged to write down a local language in Cameroon (well I stopped after my stay there and after realizing that this language is dying fast with people being unable to remember basic words like "giraffe") I stumbled over apparently "universal" concepts such as plural, possessive adjectives, verb forms, a three-person system and so on. Isolation during Coronavirus hysteria made me pick up studies on the Vietnamese language. It has plenty of literature and more or less a standard. AND it is contradictory to "universal language concepts": Everyone is third person ("tôi", mostly translated as "I", simply means "subject", example: "vua tôi" = "king and subject / subserviant", in this fixed expression the "and" is missing), no verb forms (time is guessed from the context or taken from a prefixed "new", "soon" or other words), instead of plural forms you simply say "all" or take a general plural marker ("những"), any possessivity is expressed by word order (example "mother I" = "my mother") or the inclusion of the word "của", meaning "property". Any inclination is avoided except for tonal changes when doubling words or expressions (example: "Ngày xửa, ngày xưa" = "earlier day(s), earlier day(s)", meaning "once upon a time"). The good sides of this language are its head-first principle (though interrupted by loanwords from old Chinese) and the ability to express oneself without necessarily defining time, gender or number. On the bad side are the dialectal sound mergers, double meanings and a difficult pronunciation with minor subtleties ("tay, tai, tài, thai, thay"...) hard to detect and a vocabulary hard to remember (without "Anki" cards almost impossible for an adult). *** In short: I think that isolating languages kicked "universal grammar" into the linguistical trash can.
The concept of abstract colour names is missing in more langugages than just Pirahã. When I was in Cameroon I was encouraged to learn their local language, a local accent of Éwondo with the special feature of replacing "f" with "p". When I asked for colour names the answer was "by the colour of..." There are no specific colour names except their word for "ripe" meaning red, orange or yellow. Then there are several words for black or white, their usage depending on what you describe. By the way, this same thing seems to apply to Yoruba. For example the Yoruba word for "brown" is "àwò igi", literally "colour tree". -- Rob Words has a wonderful video on TH-cam explaining how the concept of colour names evolved in English.
I recently found your channel and I love your videos! I'm new to linguistics and language history. I've never heard of recursion before, and have a question about it. To me it seems like the use of a pronoun necessarily involves recursive thinking. In the last sentence in the example, "They are the same nails," the listener needs to know that "they" refers to the nails that Dan bought and that the speaker wants brought to them. Or is there no recursion because the concept tying the pronoun to its antecedent are in different sentences? I guess I need to read up on what recursion is.
Many things about grammar are primarily because we are used to hear sentences spoken in a certain order. If I would say for instance „the window I close“ you still would understand what message I am conveying and the action I am planing to take. It only would sound odd to the listener if they are not used to a sentence constructed that way. However if the listener had been growing up in a society were everyone would construct the sentences this way, they would consider this tobe „normal“.
I think language, in the uniquely human sense, began when it became capable of VIRTUALITY. Animals communicate in the here and now. To speak of yesterday, or of possibilities, requires the ability to construct a virtual idea in the listener's mind. And that in turn opens vast new possibilities in cultural evolution through story, spirituality and myth making. That raises the question of how the Imagination evolved to want to go there. What is the imagination ? What comes first, language or imagination ?
Simon! I feel like we would literally talk for hours on end and never run out of subject matter. I'm currently reading 'The stuff of thought' by Steven Pinker. I imagine you've read it? But if not I'd recommend it. I can't speak on Pinkers conclusion, because I have yet to finish. Personally I think along the same lines as you, if I haven't misunderstood your points. Language seems to almost be an exact expression of our underlying consciousness. And what I mean is that our brain clearly works in ways that language then reflects, once it has developed. This may be a little outside of your interest, but I have the idea that 'consciousness' as we see it in a scientific/deterministic sense, is so directly intertwined with language, that I would suggest they are fundamentally the same. And what I mean is that we would not have one without the other. I'd be interested to hear your thoughts on this idea. This way, the pressures to evolve all different parts of our brain to deal with higher levels of cognition, would have just produced language as a side-effect. And that would then of course be selected for because of its usefulness. This feels right to me when we look at other apes, or 'intelligent' animals who seems to have 'degrees of consciousness', and we see them also grasping some aspects of language (but not the whole).
Hi Simon, thanks for the wonderful video. I was wondering if in your archeological studies you'd found anything, such as an artifact or evidence of a religious or other cultural practice, to suggest there may have been some aspect to an extinct culture that reflected unique grammatical features that are not found at all today? You make a great point by saying that most languages in terms of the total that ever existed are now extinct and that we don't have the opportunity to study them intensively (Thracian for example).
Regarding the poverty of the stimulus thesis, it would make sense for something as fundamental to humanity as speech to have prebuilt models, HOWEVER humans are also in general very specialised in constructing models for any new phenomenon, so it wouldn't be surprising if a child managed to model the language of its parents just by listening without specific language slots ready to receive types of words
It‘s all about types. Language is typeborn and typedriven. Types are already present in big apes, some birds, dolphins and even in octopuses. Types provide a digital pattern of information processing totally different from the prior analogue modes . Richard Dawkins postulated memes in 1976, but memes have to have a substrate, an implementation. The downfall of the Journal of Memetics in 2005 was due to this lack of a functional foundation. I think types were at first used internally for fine grained representations and movements. At a certain point they evaded into gestures and sounds. This is the beginning of language. Grammar is in its core a categorical comprehension and expression. Language consists of monoids, monads, semigroups, covariant and contravariant functors etc. So it can be described by the terms and laws of category theory. I think, it makes sense to divide language into a pure internal subsystem, that’s highly composable and insures the fluent production of language and a monadic system, that makes language expressable, for instance in speech, body movement or literature. In contrast to the language of the honey bee, the human language can be easily transformed into infinite subsets of Domain Specific Languages (DSLs). If we want to get any further in language analysis it’s inevitable to take care of the non-communicative foundation of language. Take this two videos to reason about the problem. They show emerging expressiveness of a, in my opinion type driven system, but not in terms of a classic language. th-cam.com/video/NPxgRAp0Z_0/w-d-xo.html th-cam.com/video/Wos8kouz810/w-d-xo.html
Simon has a microphone. Somebody gave Simon a microphone. They are the same microphone.
Very difficult to insult people if that's what you have to do every time. I expect the Pirahua must be extremely polite. And fit to explode with inexpressible resentment.
Someone made microphone belong Simon
So 'They are the same microphone' is just a wordy way to create a post-positioned relative pronoun.
@@teebo_fr_en_it As in, "Somebody gave Simon the microphone that he has" ?
@@MatthewMcVeagh Yes.
Simon, don't be self-conscious about the hand waving. It's going to be important when you start appearing on "smart people talking about stuff" shows for the BBC.
That would make him the first smart person on the BBC.
TED
Don’t let the BBC get their claws into you.
I pray for Simon to get his own BBC series!
Absolutely. When I was a kid, we had Magnus Pike. He was my hero :). th-cam.com/video/OL1kT3-BPuo/w-d-xo.html
This channel is such a hidden gem
It's only hidden until you find it!
@@citolero..how profound
@@SaintJames14 Well, yeah... "hidden" is not a property of the channel, is it? Just call it a "gem", that'd make more sense.
@@SaintJames14 A channel is hidden. You found a channel. A channel is not hidden. They are the same.
@@Релёкс84 thank you, sensei
Our boi here speaking Pirahã like a pro
Well, if you noticed, he still retained his non-rhotic accent. (Just joking. That was impressive as hell!)
it was brilliant! shows what knowing IPA can do for pronunciation.
Y'know, I was expecting the Pirahã to be mentioned before even clicking on the video...
Didn't expect heccin' full sentences to be breezed through like that! Even if I'm a li'l sus of the accuracy of the pronunciation but never mind that.
That was AWESOME!!!
@@impishDullahan There are a couple of phonotactic rules to bear in mind - /g/ is realised as [n] if it comes before /o/, and /t/ and /s/ are palatalised in certain environments, so that may be why it doesn't look entirely like the spelling. It's absolutely possible I've made some other mistakes, though! It's not something I really have any experience with.
That was a FANTASTIC read of piraha!!
I appreciate your video style. I get a bit shell-shocked by some of the expository videos that are frenetic, high-pitched sales jobs. The simplicity of conversation and simple word illustrations allows me to watch these videos and relax.
I agree that the simple style, girded with such a depth of knowledge is extremely appreciated. This is such a fascinating topic, but it does require a bit of thought. I think that might be why I appreciate the lack of glitz and the anti-frenetic style.
as someone who has tried taking linguistics courses in university and not enjoyed them, your videos are amazing. such great explanations in clear terms, truly has made linguistics feel so much more present in my life and easy to understand. keep it up
Couldn't agree more
There is a (sign) language that was developed spontaneously, Nicagraguan Sign Language.
There was a group of deaf kids in Nicaraguan in the 1970 - 1980s. Teachers tried to teach them how to read spoken spanish from lips. However the kids were unable to learn the language. During the breaks the kids were playing outside with themselves. They started to communicate with themselves by hand signs. Those signs evolved into a full blown sign language with it's own grammar.
The thing about Pirahã having no basic color terms is not as weird as it sounds. They do have color words; they're just the names of plants with those colors, similar to how we have "orange", "violet", and "lilac" as colors.
The thing about them not having numbers seems to be pretty much what it sounds like, though. They aren't the only culture that doesn't count, but most languages like this still have a few small numbers. That being said, Pirahã does have a word for "one"/"a few" and another for a larger number. It's probably not that similar, but it does remind me of how a German word for "some"/"a few" is "einige", which I think is related to "ein(s)", the German word for one (which is cognate with English). I've always wondered how that change happened since I noticed the connection.
It has also been shown to some extent that practitioners of languages without numbers have immense difficulty in understanding abstract thoughts. Im sure there is some debate on how related this may or may not be to the nesting of sentences.
@@Sugmatron The problem with that is that they don't necessarily have trouble with numbers BECAUSE their language lacks numbers. An equally valid conclusion is that they have difficulty with numbers because they didn't learn about them as a child. That is, the fact that the language lacks numbers and the fact that the people have problems with numbers both stem from a third factor: The culture doesn't count. This is a classic case of "corrolarion does not prove causation".
FYI, the above is an argument against what you said being evidence for the Sapir-Worf hypothesis. I know you didn't explicitly say it was, but it seemed like you were implying it.
Also, I notice you actually said "abstract thought" and also "practitioners of languages without numbers". I don't know what other types of abstract thought you were talking about, and I don't know what other languages that Pirahã you're talking about. I know that the Pirahã specifically supposedly don't care much about things that far away from them in space or time, e.g., ancient history, legends other countries, etc., unless it specifically affects somepne they know or someone that someone they know knows. The only problem is, I think a lot of native English and Spanish speakers kind of like that, so I'm not sure how Pirahã-specific that really is. I'm also just going off of what Everett says for a lot of this.
@@Mr.Nichan I agree in that correlation is not causation. But I'm not certain that one can easily separate culture and language in such a way where you can decide if it was the "chicken or the egg" so to speak.
I was not speaking directly on that hypothesis, although related in hindsight. Just the idea that the linguistic (or cultural, as much as it could be separated or attributed) lack of counting and other forms of abstractions may have some relation to the lack of nesting sentences.
I said abstract thought as there was a study years ago where researchers worked with established language practitioners of languages without numbers along with non-standard sign language practitioners in Brazil (I'm not exactly confident in the nation, I would have to find the study once more). The users of these home-made sign languages which were deaf from birth and had no access to a modern sign language, along with the standardized spoken language practitioners showed an inherent lack in understanding of abstraction. I specifically remember a focus of the study being how the practitioners of the sign languages could use money and make change as easily as the average person, but once numbers, or even time was abstracted away from physical objects or actions the understanding went with it.
Quite interesting stuff.
@@Sugmatron Those particular abstract thinking skills you mentioned were pretty quantitative. I'm not sure if there's any reason to think that difficulty would translate into non-quantitative types of abstract thinking. I also wonder how they were attempting to abstract it away from physical objects without switching languages. I suppose they used numerals and something to do with counting time?
In spanish there is "unos" for a few, like your example in German. Unos carros = some cars.
There is so much of interest to learn from these videos but just for starters lets all try to display the humility shown here. Whatever we do , lets ne a bit more like simon.
I’m from Hawaii, so it was cool to hear you mention Hawaiian creole (though we just simply call it “pidgin” here even if it’s not actually a pidgin language). It’s pretty interesting because growing up I never thought of it as a different language.
It’s often looked down upon as a lesser form of English, or broken English, and people who speak pidgin primarily (I don’t, but many of my older relatives do) have similar problems to the ones mentioned in your AAVE videos. Code switching is very common, and even among people who speak primarily pidgin there’s the mindset that it’s lesser, or uneducated, or low-class-myself and many of the people in my age group (at least the ones I know) speak standard English primarily, and it’s not uncommon for my older relatives who speak pidgin to say that we “talk good” and often older people who speak pidgin try to encourage their kids to “talk good” because they think it will help their kids with school and jobs and stuff. I admit that growing up part of the reason I preferred standard English was because it seemed more proper and educated (plus other factors, of course, like it being the primary language of education, the fact part of my family is from the mainland, and the fact that most of the media I consumed growing up was from the mainland). As a result I still don’t feel that comfortable speaking pidgin, because it feels like I’m putting on an act because I never actually spoke that way growing up, but after taking some linguistics in college and learning more about the language and it’s rules I found a new appreciation and perspective on it.
Cheers to whoever sent Simon the mic. I must admit, I thought about doing this when the phone came into play, but I thought it might be a bit presumptive. Well done whoever actually had the motivation to do it!
How Did Language Start is by far my favourite topic. It makes my brains go funny. Like bottomless chocolate. Looking forward to the next episode mate
verbal archeology
My favourite example of recursion is "Those are the only vegetables which I don't know where to find out how to plant."
Reminds me of my favourite sentence, which is an example of preposition stranding: "What did you bring the book I don't want to be read to out of up for?"
@@nickhight Love it. Another genuine utterance I've relished over the years is, "There's not no nothing nowhere." - one certainly get's the gist.
@@nickhight These sentences hurt me
@@nickhight It took me quite a few reads to understand this.
@@Leandro-bj6jh Prolly cuz you're on the interwebs and can't use your hands to defend yoursel'. ;)
First, apologies for not being able to focus much on the content, your voice per se as well as the way you articulate things is just mesmerizing.
Chomsky's whole point is to demonstrate that language is more a biological phenomenon than a cultural one, while Everrett would be the reverse. I did a bunch of Chomskyan linguistics in undergrad, and I'm pretty sympathetic to Chomsky's ideas. I think his starting point makes sense. The marked cognitive difference between humans and other primates (including humans' ability to use language) must be due to a relatively recent evolutionary event. If that event produced the biocognitive capacity for language in humans, then surely that capacity *looks* like something, right? It must be, in some way, tangible. The capacity for language that arose is ultimately biological, so it has limits, patterns, and so on, that are hardwired into the human brain, just like how any other biological system in the body and brain also has limits and patterns. And since that innate hardwired capacity governs how language arises and develops and is used throughout human history and across the world, we should expect it to exert the same constraints on all human languages. So it all sort of makes sense to me.
In saying that, since undergrad I've since become more convinced that there is a lot to be said for Everrett's camp, which sees language more as a cultural tool. There is lots of good evidence that culture drives language and that language reflects culture, which I won't go into here. Though, this cultural tool take still has to be stacked on top of some biocognitive capacity for language. I guess it's possible, as you mention, that this biocognitive capacity is just a combination of higher-order pattern recognition, symbolic reasoning, and other cognitive abilities. And the point that children may quickly pick up and make sense of grammar based on word distributions, not anything innate, is a good argument here. Thanks for making me think.
You might also be interested in exemplar theory, which says that speakers (including children) store words that they hear in the mind with all their acoustic, social, syntactic, and other contextual information. So when people (again, including children) speak, they don't piece together grammatical sentences (and develop grammaticality intuitions) based on anything innate, but rather on the wealth of exemplar words and sentences that they've amassed.
I don't think it's completely fair to divorce the phenomenon of language entirely away from its biological substrate. However, Chomsky's single-step hypothesis is far-fetched, in my humble opinion.
While a gain of function mutation to the coding region of the ancestral variant FOXP2 could have certainly accelerated the development of grammar in ancient humans, mutations to genomic regions that regulate how FOXP2 is used by neurons could have an equally if not more potent effect on the development of neural circuitry for speech. In fact, the two mutations in human FOXP2 previously thought to be responsible for human verbal skills are shared by Neanderthals. However, there is a mutation in a regulatory region of the gene that is present in humans but not shared by Neanderthals and Denisovans. I think it's more likely that several moderately advantageous mutations had accrued in the ancient hominin lineage that could confer language, but humans have managed to acquire (at least) one more that made learning and systematising language more efficient. However, either hypothesis is incredibly difficult to prove.
It does seem like the trail goes cold once you try to actually describe the biological basis of a universal grammar. Exactly which proteins and genes are responsible for language? Chomsky's UG is, at the end of the day, based on a biological assumption. Most of his work is not about biology; it only uses the biological assumption as a foundation. From there, Chomskyan linguists launch into the way in which grammatical sentences must be generated in the mind if there is some underlying UG.
@@nickhight I agree. Chomsky is just using biology as a premise for his argument. But that didn't stop a bunch of biologists from trying to test those assumptions though! I think it's an exciting field.
@@nickhight While that is true in a way, if you look at Syntactic Structures, Chomsky's first book, you see that what he is attempting is a mathematical rather than a biological approach. You have to remember that he was reacting against the Behaviourism that was widespread at the time, so I see his theory as a specialized theory of behaviour, and to that extent divorced from genetics and mainstream biology. The point of contact would be neurophysiology and neuroanatomy, really, rather than genetics.
Does it have to be an evolutionary event though? Couldn't you just argue that the social nature of humans (and other primates) the strong ability to teach can basically "snowball" the development of a common tool such as language? I think you can apply this same line of thinking to prehistoric tools and even see it in action today with new tools such as smartphones and other technology, there is a very strong ability to pass knowledge and understanding down, and this dissemination of knowledge fundamentally makes primates prone to adapting their behaviour in ways that exceed the speed of genetic evolution because the person learning from another doesn't need to put in the same effort as the person who figured it out in the first place.
I get the sense this young man’s recordings will be listened to a hundred years from now.
Fascinating. You’ve made the uninteresting interesting and yes, I do miss your weirdo naturalist excursions. More toads please....
Very strong evidence for innate universal grammar is provided by the case of Nicaraguan sign language. During Nicaragua's long civil war the education system for deaf children collapsed, so a generation of them were just brought up isolated at home, and also the teachers had retired or were dispersed. So when peace was restored they decided to provide education for deaf children, and collected a number of them together. However, they needed to train the teachers, and decide what system they were going to use. So at first they left the children alone while they discussed these matters. Amazingly, these young children, who didn't know each other, invented their own language, a signing language. This was so successful that it has now been instituted as the official deaf sign language for Nicaragua. Analysis shows that as with other signing languages, the grammatical structures are the conventional ones you might find in a spoken language. The idea that this was generated by culture seems hard to justify or accept.
Congratulations on the channel, by the way.
Michel Dansel, under the pen name Michel Thaler, hated verbs so much he wrote a whole novel without using any. He still manages to express his meaning quite comfortably. But I suppose the point is that he had to work really hard in a language that already prefers nouns to verbs to contrive a way of doing so. en.wikipedia.org/wiki/Le_Train_de_Nulle_Part
You're very modest. I personally believe in UG - but I do think that the evidence for its existence is theoretical. That is, it is a logical construct. If there were not a language acquisition device in a child's mind, we would expect errors that do not occur. From that I do think that the binary branching model is the best we have for how language can plausibly be processed in the brain. Even if our brains operate just like Google translate in every respect (for instance), then surely that is "universal grammar" as well. It would be an overhaul of the Chomskian binary branching model, but it would still be part of the biological endowment. Universal grammar is the scientific name for whatever it is that helps us acquire language, and is specific to human language (rather than an alien one). I hope this makes sense and that I haven't presumed anything incorrectly.
I always loved Chomsky's activism, but universal grammar has always appeared to me an unbelievably bold a claim given the available evidence. Science proceeds because of the certainty granted to it through the poverty of it's epistemological means, and not by the explanatory attractiveness of it's hypotheses alone.
I feel pretty comfortable not developing even a tentative opinion on this question, partly because for some of the points you described I can hardly imagine what the difference would actually look like.
Just one very small nitpick. "Who did Jack ask when Jill saw?" can, as wildly unnatural a construction as it is, be grammatical. The verb "to see" is technically ditransitive. So, we could say this if, for example, Jill had been blind most of her life and saw for the first time. Here's an example of what you're referring to: "Who did Jack ask when Jill inflicted?" This sentence should immediately provoke a "what?" response because you can't rationalize it no matter how hard you try. Something needs to be inflicted, whereas one can simply see without necessarily needing an object. I'm hoping that the response to this comment, if it gets one, is an "I see." ;)
Wow. Free mic. That was really cool. There's still hope for humanity.
I actually found a couple of linguists online (including the youtuber, Simon) while researching recursion...nice one.
I very much appreciate the slides! If there's an auditory version of dyslexia, I think I have it. Reading along helps me so much!
It's also nice to see your face, too, so it feels like a conversation not a lecture. And the nature shots are beautiful and peaceful.
Yay for balance!
It’s called auditory processing disorder, your auditory dyslexia. I agree that a mix works best.
My university is very Chomskyan, so I am very biased, but I find the fact that damage to certain parts of the brain (Broca's and Wernicke's areas, to be precise) leads to aphasia, certain parts of language being damaged (language production in the case of Broca's aphasia and language comprehension in the case of Wernicke's aphasia) a strong argument in favour of language-specific areas in the brain.
I've gotta say, this is my favourite channel at the moment. Look forward to each new upload. Thanks for the great content!
I remember reading an interview with Chomsky in Omni magazine about forty years ago. The interviewer kept asking him how were were going to communicate with space aliens when we make contact. Chomsky quite reasonably said he had no idea because he didn't know what the aliens would be like. Chomsky also casually mentioned the term "universal grammar" as if it were assumed the interviewer was familiar with it--which led to another round of questions about aliens. Chomsky corrected him, saying universal grammar was only universal to humans. The interviewer replied, "Maybe we shouldn't call it universal, then."
I have a layperson‘s fascination with linguistics (and specifically Old English), and you make it so much easier to understand than other TH-camrs.
The videos only get better as time goes on. Can’t wait to see what the new stuff looks like.
Just in case anyone else is learning German like I am, remember that the 9:05 sentence is very literally translated and therefore doesn't really work. It would have to be 'ich zerbreche...', 'ich zerschlage', 'ich zerstoere', 'ich mache das Fenster kaputt' and so on. The zer- is necessary and important as it's a verbal prefix that suggests some sort of dissolvement or destruction. Very interesting video though! Keep them coming!
yes puking came to mind.
"Schieß dem Fenster!" - "What?" - "Shoot the glass!"
Nice video, Simon. In this kind of discussion, it's important to keep clear in our minds two distinct issues: whether humans have an innate capacity for language; and whether there is a universal syntax. Some of the commentators in this board seem to be confusing the two questions. I find Everett's argument convincing, namely that there is no universal syntax. Children do seem to have an innate ability to quickly deduce linguistic patterns (rules) associating sounds with concepts and combining words to make sentences. But the rules themselves seem almost completely arbitrary (i.e., cultural); except to the extent that they must accommodate universal human cognition.
I like how you just casually speak Pirahã in the video. Awesome content Simon!
Props to whoever got him the mic! And nice work, as always, Simon!
I definitely lean much more towards the Chomskian side. So I disagree with you on that.
I'm sorry, I hope you don't hate me for it.
What convinces me the most about it is the fact that language as a thing is so universal. Like, the Pirahã language may be very unusual, but it is still a language. Where a population is hearing, there will be an oral language where sounds are made with their mouths to represent individual semantic concepts that are later placed linearly next to one another in extremely complex and specific ways to encode meaning. And the same will happen with a sign language if the community is non-hearing.
When you say that linguistic universals may be a product of other cognitive processes that just naturally coalesce into a specific way of using language, I see where you're coming from, but the difference is in just how incredibly quick and unconscious the acquisition of language is. Other cultural phenomena require a certain degree of active learning: a child will not learn to bike, to read, to sew or to play the piano if there isn't someone who is telling them what to do and how to do it at the early stages, even if they get to master that activity and do it like it's second nature, they had to observe diligently at the onset of learning. Some kids may do some of these things without initial guidance, but not all of them. Language, on the other hand, is clearly engaging some of those cognitive processes, that may be more general, in a very specific, goal-oriented way, to force a child to acquire language simply from exposure, and the stages of acquisition are surprisingly consistent.
If our brains were just good at certain cognitive processes, and we use those things to acquire language, without a specific module that specifically aims those processes to the acquisition of language, I feel like we should see MORE variation. I know that few things are truly universal, and language is clearly formed from outside stimulus that is culturally determined, but even if some things aren't universal, I just couldn't convince myself that they are as "obvious" as to appear in so many languages that have nothing to do with one another, and haven't been in contact for millennia.
I encourage you to look into conlangs, which are constructed languages, and to see the incredibly different ways people have come up to encode meaning, that you just intuitively know couldn't be possible in a natural language. I especially recommend Ikthuil, Lojban, Kēlen, Fith and Rikchik. When I think of these conlangs and compare them to natural languages, I can't help but wonder why natural languages just don't evolve to allow more levels of complexity, like Ikthuil, or why they don't organize predicates like Lojban, etc. And I think it's partly because we don't "just use" what our brain has to offer when we acquire language, we use what it has to offer in a particular way, which encourages and allows certain things and disallows others; otherwise you'd see more diversity.
But that's just my opinion. Perhaps those "obvious" ways in which to use our cognitive abilities really are obvious, and that's why we tend towards them, and maybe our languages really are the most effective ways of imparting information, and they have these things in common simply because they're the most direct way to do it with the tools we're given. I'm just skeptical of that.
I think the biggest "proof" for chomskyism is how sign languages arise naturally, even among illiterate people-i.e people who have theoretically no exposure to language, manage to use the tools at their disposal to create meaning, to translate there thoughts into expression.
Well reasoned and compelling.
Ah, a simple short disagreement, fair enough
*Clicks read more*
OMFGGGGGGGGG
Jokes aside, great argument. I'm gonna play devil's advocate to both arguments though cos I don't know enough to make points of my own lol.
When Simon said part of his reasoning for not siding with Chomsky too much was because a language 'module' doesn't appear to be strictly necessary, my physicist mind was delighted because that's how I like to think of things whenever possible. However, I don't think that argument is particularly relevant here because evolution isn't a perfect science, there are countless examples of over-evolution and 'dead-ends' etc and with how important and common language is to humans its not a stretch to think we might have one.
Despite that, I do see a flaw in you suggesting that there is no, or little, 'active learning' and 'diligent observation' in regards to language, which I have inferred to mean conscious, logical, observation-based thought. From what I've heard we start 'observing' language from in the womb and are clearly constantly exposed to it afterward. I think its fair to say that counts as extremely extensive observation. As far as I understand, our brains, especially when so young, are extremely malleable to stimulus. While having a brain region for this this function seems to make sense as an argument why this happens earlier with language than your other examples, if you think about what stimuli are present to a baby/toddler speech is one of the biggest and so it follows that the baby will spend it's time and thought on puzzling out those stimuli first. This hinges on whether you deem a baby capable of conscious thought, I would argue yes, in a different way to adult thought but still concious nonetheless, but I'm not a neuroscientist and don't have evidence to back that up for or against. So I know this isn't a flawless argument by a long shot but I thought it was worth mentioning to provide some counter to what you were saying.
Kind of ignoring all of that and going back to basics though, I thought neuroscience had already decided we do have areas of the brain for language? I assume the Chomsky/Everett discussion must be about the specifics of their function then and how developed they are? Like whether it extends to things such as a natural distinction between nouns and verbs talked about in the video or am I completely wrong?
I’m not a scientist by any kind, just a passionate by linguistics, but I want to point few things: children until few years don’t learn the language in a way that an adult does it - their brain learn by mimic, while an adult brain lost partialy this ability; that’s why is so easy to learn new languages at early age (I have such an example in my child, comparing to myself); I make a very clear distinction between conscious speech and simply speech - in my humble opinion most of the people have a simply speech, making a lot of mistakes and having a limited vocabulary; I truly believe we are living computers, but even so I always wonder how grammar appeared! It is a concept so complex that I doubt it is human! 😁 Simon himself seems to be from onother planet 😄 joke aside, I admire him and I enjoy every video he posted ❤️
@@sinsemilia70
Agree, you’re not a scientist by a long stretch.
Fascinating stuff and nice to hear about language development in the Uk etc ... amazing ! Thankyou x
I love your nature's shots! Everytime it's so spontaneous and mystical ! :P
It's a good topic ... I don't know if languages is human based, i think we definitely have a place in our brain ''Made or evolved to be'' for ''verbal languages''. But if only a dog or a cat could have a human's mouth and vocal cord..
I would be sooo curious to see it talking because i have cats and sometime they respond and talk to us like we do. It's mind-blowing!
A pretty good question would be ; How well do you think your dogs/cats understand you?? I think we'll be shocked!
And it could maybe answer some of the languages origin's questions
Sorry for my grammaticals faults , i am a Canadian French...
I should not say that, i'm gonna be beaten down xD
First love from Iraq 🇮🇶❤️❤️
Thanks, Simon. Excellent video. Cheers.
As I have grown older, I find that I'm a frustrated linguist. As a wee lad playing in my sand pile I wondered about language. I may have just heard the story of the Tower of Babel. I was interested, even at that young age, where it began, what it sounded like and how did it become my vernacular. I wish I had realized I had this interest when I could have done something about it. I really appreciate your videos.
Thank you! I'm sure it's still possible for you to do something about it, even if it just starts as a hobby :)
i havent watched your videos in months (no particular reason hahah, ive just been busy and i like to listen to music or more mindless stuff like livestream VODs while i work) and these "how did language start" ones are my first few back--i cant believe id forgotten how calming and informative and wonderful i find your videos!! thank you for making this series, i do lots of worldbuilding for fun and your channel is a great resource for my art c:
Pirahã is so interesting.. 🤔
Cheers from Mexico, Simon 🖐
Thanks for your videos, Simon~
Greetings from Russia
Really fantastic stuff. The best overview of the subject for newbies and intermediates in this realm. Glad to see you read Don’t Sleep... I would be interested to see a video just in that book.
good for you with the mic, however I also enjoyed the lo-fi diy approach :)). Thanks for the content!
Interesting and educational. Thank you!
I like the slide heavy vids because it aids with digesting the information but I am also a fan so feel you have to express your work in a why and style that represents your vision for the channel
You're saving my sanity with your content. lol
I've said it before and I'll say it again now
I think your "B-roll" footage is really really good
Thank you very much indeed for making, this was another unreasonably interesting topic and video.
When I started to study linguistics at university, I expected to be told of systems of communication that did not conform to universal grammar in contrast to human languages. After all, a good way to understand a concept is to
consider its opposite. I didn't hear of any examples. I can think of some possibilities myself; mathematical notation, computer languages and, perhaps a good one, bidding systems in bridge.
What exactly do you mean by "mathematical notation"? If you mean logical notation that is used to express concepts in mathematics (like S-PL for example) : those are basically artificially constructed langages, with very strict grammar.
And with regards to "computer languages" do you mean programming languages? Because those fit into a lot of definitions for universal grammar too.
@@shiniri2647 By mathematical notation I mean the ordinary symbols used to express formulas and equations. By computer languages I mean programming languages which have a syntax very different from natural languages.
I have used many of them, I am a retired programmer, and found they are not like human language. The only possible exception is COBOL whose designers intended it to be written and read in something like English. This idea was quickly found not to work because programmers complained it became too verbose so that simpler non-English like structures were introduced.
@@thomaskember4628 I'm in no way qualified to comment on this, since I've never studied computer linguistics, only regular linguistics, maths and a couple of programming languages, so what I'm saying is based entirely on what I made up myself, and cross-application of my knowledge of those fields. But I have found that the programming languages I know have similar syntax and semantics to mathematical logic (meaning the classical first-order) which in turn fullfills the definition of a universal language by Chomsky, namely that there has to exist a finite set of rules for the organisation of the language. Where exactly do you think the difference lies then ?
With Daniel Everett you mention something that originally started my interest to go deeper into linguistics. Just funny to have made this circle for me ;)
And I just recognized a very interesting linguistic feature of your videos (and probably other videos on youtube): The very interesting mixture of the poles of the conceptual continuum of orality and literacy of the language of your videos. What really triggered me here was the train, when the rather scripted section was interrupted by "wait, there is a train coming". Former colleagues researched on what effect that might have on students when a teacher uses language more on the side of orality of the conceptual continuum or on the side of literacy. The quick answer: It is not that simple! :)
Language seems to be derived motorically from a more general movement feature, the „Willkürbewegung“ studied by the ethologists. It’s functional polymorphism. The evolutionary trends converge to language independently in parrots, corvids, toothed whales and primates. But it’s more sensible to view at language from a teleological perspective at the beginning. Language is mainly a feature that allows to instruct a conspecific of a precise to do. Honey bees are a model organism for a simple language. Their waggle dance language delivers a decoupling from spatial and temporal constraints imposed by pheromonal or tactile information „techniques“. When a corvid raised by hand tells you „Stay away!“ it‘s an imperative approach . When it says „ I‘m hungry“ it‘s an declarative one. All appetitive behavior is structurally more complicated than avoiding. I personally think, that language is a „child“ of the optical sense, because this form of perception is not restricted by short informational events like acoustic, tactile or olfactory. The „mild omnipresence of light“ (Novalis) opens up a categorical space, that‘s nearly boundless. The german term „Bedeutung“ encompasses such a new category. „bedeutungs“- sensitiveness has been studied by Auguste von Bayern in jackdaws. These corvids are able to reason about the intents of their conspecifics by following their line of gaze. This is one category of indirectness, that‘s only possible under the conditions of optics. Others are „option“, „expectation“ and „objectivity“. The nervous system therefore is solving differential equations, like Sabine Hossenfelder pointed out in a talk on TH-cam. I think it makes sense to combine linguistics, reasoning about the capabilities of an optical sense and Category Theory for a deeper understanding of language. By the way, the decision between verb and substantive is not a linguistic trait, but a universal one between type and function. „We routinely use types to express our high level intent of our programs“ (Nadia Polikarpova in „Type-Driven Program Synthesis“).
A huge number of linguists would argue for a genetically-encoded, universal grammar. A huge number of linguists are wrong. They are the same linguists.
Lmao
If I could be allowed to show off one more time; the sentence: "Where John had had Mary had had had, had had had had the examiners' approval," is perfectly grammatical and semantically straightforward in English.
But that's cheating a bit isn't it. If "had" is plums and "had had" is prunes, prunes pluperfectly got the nod.
What's impressive is that most people are able to see through the trickery and understand it straight away, especially if it's spoken, which I suppose is testament also to our expressive vocal ability.
@@frogindeed no, not cheating a bit - and 'had' isn't referring to anything other than had.
This is my first time seeing this sentence, and as a native English speaker I'm not finding it very straightforward at all. Could you expand and explain what the meaning is supposed to be?
@@baronOdaighre Sure. Two people, John and Mary, have completed an English exam. In one sentence John had written, "Sid had some bananas", whereas Mary had written "Sid had had some bananas", which the examiners preferred. Therefore where John had had Mary had had had, had had had had the examiners' approval,
If I had had a bit more intelligence, I might be able to understand that sentence.
It's interesting that you brought this up, Simon. I've been teaching myself Ukrainian and sometimes I will have to choose the missing word in a lesson and I don't know any of those words. First, I see if the word is probably a verb or a noun, for example, and then I look at words I do know and see if the new words are spelled similarly, like maybe it's the plural form of a word I know or something related to something I already know. Sometimes I just shot in the dark guess. 🤷♀️But it's interesting how all languages share some very basic rules with each other and even if you don't know the language, you still have some basic knowledge of sentence structure or that verbs have different forms.
Very nice video, as usual. For those interested in the subject, I'd recommend two older pieces (in English) that, without exactly arguing for or against universal grammar, offer insightful critical remarks on the frameworks from which the topic has been debated. These are Eugenio Coseriu's essay "Universal (and other) universals" (available at www.romling.uni-tuebingen.de/coseriu/publi/coseriu128.pdf) and Ernst Pulgram's review of Chomsky's 'Language and mind.'
We think recursively because every verb and noun in our inner dialogue comes loaded with associations, memories and emotions; in order to form a spoken sentence we must submerge all but that which serves the intent of the moment. All else is, as it were, put in parentheses of silence.
I’m no expert, but the “arising from human thought” perspective sounds like a lot of things I’ve heard ascribed to Chomsky, eg “language is just human thinking translated into sound”, and “human thought is symbolic and grammatical”. Am I missing a part of the argument here? Really cool video and interesting topic btw.
I think you may have a point in that there is a school of thought which says that human "rational" thought is based on human language. The "arising from human thought" argument that he is discussing in this video, on the other hand, is more of the idea that human language is just one application of human thought, which is not fundamentally linguistic. The direction of the causality is different.
He only briefly touched on the viewpoint you mention when he discussed that some people seem to "think in language" constantly, some apparently don't really ever, and most people are somewhere in-between.
Given that we're born able to mimic the rhythm and cadence of our mother's tongue (our 'accent') it seems we have a universal 'music' module, but acquire our grammar as we do our phonemes: by fitting them into the tune by trial and error. news.bbc.co.uk/2/hi/health/8346058.stm
Very well explained - thank you.
Isn't there a lot of neurological evidence for universal grammar, namely TBI in different regions or strokes in different regions either causing severe language impairment or no language impairment?
Hey Simon. Love your videos. I Was wondering if you have any of your content in podcast format? Would love to listen on the go..
Fourth love from Australia.
Have you read John Truscott’s Consciousness and 2nd Language Learning? He makes an interesting argument for the dedicated language learning module idea.
Or if we want to consider rational processes vs. UG, then what can we make of English attributive adjective ordering? The general “rule,” as I understand it, is that adjectives more intrinsic to the noun go closer to it: “big green tree,” since trees come in various sizes but are generally (culturally?) perceived to be green. If we are presented with “green big tree,” our brains inevitably scramble to find a context in which this makes sense, such as some bizarre horticultural experiment with a row of small varicolored trees and a row of large ones: Since the speaker would need to distinguish which row he/she was referencing, you might get, “But what about the green big tree? It’s doing nicely, don’t you think?”
Thank you for your videos. I greatly enjoy your subject matter and the clarity you bring to it.
love this channel!
Based on how evolution works, I think it's far more likely that language evolved out of a collection of programmes in the brain and not from a singular, discrete programme. Much of the architecture of the brain is deeply conserved across vertebrates. This is clear not just from comparative neuroanatomy, but the central nervous system across vertebrates have similar patterns of development from embryonic tissues. Nature tinkers with extant gene expression programmes to build more complex emergent properties if it is so advantageous. Moreover, neurons have special evolutionary pressures on them, and thus have additional mechanisms to generate diversity in gene expression patterns to make such elaborate functions robust (e.g. alternative splicing, stop codon read through, 5' UTR regulation, etc.)
Edit: Clarifying that this is mostly my perspective as a person who studies genomics and the evolution of neural circuits (I do not have expertise in evolution of language or neurolinguistics per se).
Re early childhood learning: young children can and have learned a wide range of astonishing skills, when their environment has been rich in that particular practice. Musical instruments, painting or drawing, other craft and technical skills which are distinct in many ways from spoken language. Is there a piano machine in the brain? An African drumming machine? A pen and ink illustration machine?
No there is not. There are a set of resources and structures in our brain and body which have historically accomplished an amazing array of techniques and crafts, and can clearly be harnessed very early, shaped almost entirely by what matters to the humans around the person, and what they have to offer materially.
I just don't see how fast achievement in language (also not universal of course, though nearly so) is evidence of a structure uniquely human and uniquely for language.
Also perhaps worth questioning whether Chomsky's idea A: lives in a larger project of making sciences "harder" to increase their status, and B: has an unfortunate association with some nasty ideas about disability and neurological difference, which have emerged from the great man's mouth on more than one occasion, and been backed by his followers. I have twice had dime store Chomskyianism thrown at me as an argument against supporting communication strategies for non verbal adults, so that's fun.
The point is that a sociolinguistic view of language propogation is not only probably more correct, it is also more humanistic and fun and full of life, less reductive, more multiplicative.
For whatever it's worth to you, I think your face is a perfectly nice thing to have on screen to accompany your mellifluous voice. No nature scene can compare. ◡̈
3:25
Technically this sentence could make grammatical sense, if say, Jill had been blind but was suddenly able to see again, and Jack had gone to ask someone about it. Of course, this completely changes the meaning of the sentence.
Just to make the point, Chomsky himself clearly distinguishes what he meant by the term (which he concedes was poorly named) "universal grammar" - essentially the fundamental observation you cover at the beginning of the video - and the field of divergent theories that spawned from this base observation. In his book "What Kind of Creatures Are We?" he goes into a lot of detail about this, and also notes that he considers a few of these theories more plausible than the others but isn't "married" to any of them and has discarded a fair amount of his own work in the field based on new evidence (generally how scientific inquiry is meant to work), but that the basic observation stands. So it always strikes me as odd when people argue that "Chomsky is wrong on universal grammar" when what he meant by it is so different from, far narrower than, and comparatively uncontroversial to what is actually argued against, and what is argued against is itself usually a strawman version of the multiple theories in that field.
On the evolution of language scale, I agree those are somewhat extreme views. I think I lean a little more to the Chomsky side. I don't think there is a single language handling brain module, because there are at least a couple (maybe three or four, I don't remember exactly) different areas of the brain that handle language related functions, but also the portion of the brain that handles the articulation of the mouth, tongue and larynx must particularly developed to allow for language, which blows the other guy out of the water.
I don't know how long ago the Chomsky position was formulated. It would have been reasonable in the 70's, but we've learned so much about the brain in just the last twenty years that the position isn't tenable. I would be curious to know if he has a different position now. Mostly because he says a lot of clever sounding things that are quite daft, and I wonder if that extends to his actual area of expertise.
Can you explain more about the gaps in the argument that the lateralisation of the brain led to human language? I haven't read much about this topic since Julian Jaynes wrote about it a while back.
Very enjoyable as always
This is so interesting Simon
Sabes si hay canales similares en español? Sería interesante ver videos sobre la evolución del español también.
@@JonnaaM Hola! me imagino que tiene que haber. Yo te diria que busques bibliografia sobre linguistica de las universidades - son gratis por lo general.
@@ohlamaria697 Buena idea! Soy un noruego aprendiendo el español, me interesa saber más de esa evolución
Audio sounds wayyyyy better man.
We can actually hear the fundamental frequency of your voice now!
I totally love your presentations, Simon. Although I certainly support your interest and trajectory in archeology, having an abiding interest in Greco-Roman Antiquity, I also think you’re a natural for linguistics or consciousness and theory of mind studies, which appear more and more to be inextricable from the evolution and study of language.
Thank you! :) Great to hear that kind of validation from someone interested in antiquity
I have a gut feeling that the study of machine learning, in particular language models, will help solve this mystery. Because neural networks do bear similarities to real brains, and in some cases are essentially blank slates full of neurons, inputs, outputs, hidden layers, working in conjunction with other neural networks, and they do acquire emergent properties, in some cases literally evolve them. Similarly, existing language models, such as GPT-3 can become capable of other types of abstraction, for instance being surprisingly good at arithmetic.
However, the assumptions that underlie the language models are not neutral. So, rather than solving a mystery, the result might be sophisticated confirmation bias.
People have been proposing the same thing for about fifty years and there doesn't seem to be much progress. The machines get better and better at imitating written language in particular, but it's a bit like the assumption you can train apes to talk using sign language. Looks like a good idea, but when you try it it doesn't work.
It's an interesting prospect, I lean towards the confirmation bias theory. Because although we can jokingly say some languages can feel like maths, Latin springs to mind, they're generally not so mechanical in nature.
@@christopherlord3441 the one difference (use of AI versus trying to impart human communicative capacity to an ape) is that you can attempt to approximate the cultural impact over time of developing and passing on novel methods of communication that then overt time get reinforced , adjusted, or abandoned. Not that AI might not be a dead end, but I wouldn't rule out it's potential future role in testing hypotheses of language development. It reminds me of the shift in computational in brute force chess analytics like IBM Blue, vs the AI generationally self learningethods of AlphaZero. I'm hopeful sound scientifically rigorous use of AI can sid in adding insights in this field.
@@user-eq8ww1gr6v Yes I expect so but analysis of neurophysiology and neuroanatomy plus modelling of real human language will get there first in my opinion. You might be right.
Great Video, as always. Thank You Simon!
I've never seen any evidence that animals don't have language. Or that they don't use sounds to communicate. One time I was walking to work and a baby bird was on the ground, meanwhile the whole flock of related birds were freaking out about this baby bird. If you tried to get near it they would dive bomb you. They were chattering and chirping like it was the biggest tragic story since the Hindenburg.. I'm sure one of the things those birds said would translate into english as "Oh the Birdity!!"
The question would be whether or not the sounds are reproducible without the external stimuli. That is to say, when a person laughs, there can only be on meaning to that sound: it has an intrinsic value which cannot be "faked" in the sense of it always denoting a particular emotional state. Human language however assigns sounds abstractly, an often arbitrarily: the sounds cannot be relied upon to have the same meaning regardless of circumstance, and thus requires speakers to know which sounds are being applied to which concepts. A human scream, laugh, shout, etc. are not arbitrary, and have intrinsic meanings, which is what most other animals' vocalizations achieve as well.
@@mr.flibblessumeriantransla5417 Vocalizations aren't the only way to communicate. There is sign language for instance among us humans. Furthermore, for instance among dolphins they are able to communicate sound pictures to each other. For instance, say a dolphin uses his echo location to detect an odd object like a crash plane in the ocean. It can reproduce the auditory clicks that correspond to the sonar picture of the crashed plane. Essentially they use sound to make pictures they process visually and they can communicate those echo signatures to other dolphins essentially forming a picture of the object in the minds of others. That's communication. That's language.
@@malrec I didn't mean that there were no instances of other animals using what could be classified as language, just that its very uncommon among most species to produce vocalizations that aren't merely representing emotional states. Dolphins, as you mention, are a great example of another species which creates new sounds and assigns them specific meanings, hence: language.
As for your OP, it would depend on the species of bird. Most wouldn't have the cognitive capabilities to formulate sounds and assign them abstract meanings, though some of the more intelligent species likely could.
It's also worth noting that "language" and "communication" aren't necessarily synonymous. Language relies on the ability to convey information within a predetermined framework, which requires both the speaker and listener to agree upon the chosen sounds. This is different from simple vocalizations which express emotional states, like crying, shouting, laughing, screaming. While there's no precise "line" dividing "language" from other vocalizations, the point is language uses sounds that are inherently _unreliable_ in that if the recipient doesn't also agree upon the application, then no communication can occur. This means that language development likely goes hand-in-hand with cooperation, with greater sociality creating better opportunity for ever more intricate meaning to be expressed.
@@mr.flibblessumeriantransla5417 This is very well said and probably a much better response than I deserved. Thanks Flibbs.
My dog has different bark sounds for "someone that doesn't belong there is outside the house", "there is definitely something dangerous outside the house", "uncle has come home, I don't like him but unfortunately he lives next door so he may pass" and "dad has come home and is outside the house" followed by "please open the door for me, so I can greet him and make a fuss"... This is just the translation in English, because she barks in Italian. By the way, I'm German and I discovered so many things that are impossible to express in Italian. It feels as if they lack half of the words I would like to use. But, it might probably be the same the other way round. I find it fascinating how language forms our thinking and causes or expresses different mentalities.
Music; singing developed into language and yes the brain cells linked to sounds and notes are older, evolutionary speaking, than the language center cells. Plus we've found bone flutes that predate when language was thought to have developed.
I like your videos with slides inside them and I cannot lie
Nice to see Chomsky, something that applies to computer scientists as well!
11:50 This reminds me of stack-based programming languages (which includes most assemblers) or reverse polish notation, where instead of (1+2)*(3+4) you'd write 1 2 + 3 4 + *. This eliminates the need for parentheses as the order of operations is unambiguous.
I've been reading Peggy Mohan's Wanderers, Kings, Merchants: The Story of India through its Languages, Penguin / Viking, 2021. Although Mohan writes in a way that's accessible to non-linguists like me, I still paused midway to look at your video as it covers some similar ground, in order to understand the subject better. If you do have the time then please review the book I mentioned. I, at least, would certainly read it very avidly.
Another thing about "universal grammar": Being encouraged to write down a local language in Cameroon (well I stopped after my stay there and after realizing that this language is dying fast with people being unable to remember basic words like "giraffe") I stumbled over apparently "universal" concepts such as plural, possessive adjectives, verb forms, a three-person system and so on. Isolation during Coronavirus hysteria made me pick up studies on the Vietnamese language. It has plenty of literature and more or less a standard. AND it is contradictory to "universal language concepts": Everyone is third person ("tôi", mostly translated as "I", simply means "subject", example: "vua tôi" = "king and subject / subserviant", in this fixed expression the "and" is missing), no verb forms (time is guessed from the context or taken from a prefixed "new", "soon" or other words), instead of plural forms you simply say "all" or take a general plural marker ("những"), any possessivity is expressed by word order (example "mother I" = "my mother") or the inclusion of the word "của", meaning "property". Any inclination is avoided except for tonal changes when doubling words or expressions (example: "Ngày xửa, ngày xưa" = "earlier day(s), earlier day(s)", meaning "once upon a time"). The good sides of this language are its head-first principle (though interrupted by loanwords from old Chinese) and the ability to express oneself without necessarily defining time, gender or number. On the bad side are the dialectal sound mergers, double meanings and a difficult pronunciation with minor subtleties ("tay, tai, tài, thai, thay"...) hard to detect and a vocabulary hard to remember (without "Anki" cards almost impossible for an adult). *** In short: I think that isolating languages kicked "universal grammar" into the linguistical trash can.
TIL that thete are giraffes in West and central Africa.
The concept of abstract colour names is missing in more langugages than just Pirahã. When I was in Cameroon I was encouraged to learn their local language, a local accent of Éwondo with the special feature of replacing "f" with "p". When I asked for colour names the answer was "by the colour of..." There are no specific colour names except their word for "ripe" meaning red, orange or yellow. Then there are several words for black or white, their usage depending on what you describe. By the way, this same thing seems to apply to Yoruba. For example the Yoruba word for "brown" is "àwò igi", literally "colour tree". -- Rob Words has a wonderful video on TH-cam explaining how the concept of colour names evolved in English.
Fascinating I felt my IQ go up by 100!!
Same! Now my IQ is 100
I recently found your channel and I love your videos! I'm new to linguistics and language history. I've never heard of recursion before, and have a question about it. To me it seems like the use of a pronoun necessarily involves recursive thinking. In the last sentence in the example, "They are the same nails," the listener needs to know that "they" refers to the nails that Dan bought and that the speaker wants brought to them. Or is there no recursion because the concept tying the pronoun to its antecedent are in different sentences? I guess I need to read up on what recursion is.
I love words and how languages describe
Many things about grammar are primarily because we are used to hear sentences spoken in a certain order. If I would say for instance „the window I close“ you still would understand what message I am conveying and the action I am planing to take. It only would sound odd to the listener if they are not used to a sentence constructed that way. However if the listener had been growing up in a society were everyone would construct the sentences this way, they would consider this tobe „normal“.
I think language, in the uniquely human sense, began when it became capable of VIRTUALITY. Animals communicate in the here and now. To speak of yesterday, or of possibilities, requires the ability to construct a virtual idea in the listener's mind. And that in turn opens vast new possibilities in cultural evolution through story, spirituality and myth making.
That raises the question of how the Imagination evolved to want to go there. What is the imagination ? What comes first, language or imagination ?
Simon! I feel like we would literally talk for hours on end and never run out of subject matter. I'm currently reading 'The stuff of thought' by Steven Pinker. I imagine you've read it? But if not I'd recommend it. I can't speak on Pinkers conclusion, because I have yet to finish. Personally I think along the same lines as you, if I haven't misunderstood your points. Language seems to almost be an exact expression of our underlying consciousness. And what I mean is that our brain clearly works in ways that language then reflects, once it has developed.
This may be a little outside of your interest, but I have the idea that 'consciousness' as we see it in a scientific/deterministic sense, is so directly intertwined with language, that I would suggest they are fundamentally the same. And what I mean is that we would not have one without the other. I'd be interested to hear your thoughts on this idea.
This way, the pressures to evolve all different parts of our brain to deal with higher levels of cognition, would have just produced language as a side-effect. And that would then of course be selected for because of its usefulness. This feels right to me when we look at other apes, or 'intelligent' animals who seems to have 'degrees of consciousness', and we see them also grasping some aspects of language (but not the whole).
Hi Simon, thanks for the wonderful video. I was wondering if in your archeological studies you'd found anything, such as an artifact or evidence of a religious or other cultural practice, to suggest there may have been some aspect to an extinct culture that reflected unique grammatical features that are not found at all today? You make a great point by saying that most languages in terms of the total that ever existed are now extinct and that we don't have the opportunity to study them intensively (Thracian for example).
Regarding the poverty of the stimulus thesis, it would make sense for something as fundamental to humanity as speech to have prebuilt models, HOWEVER humans are also in general very specialised in constructing models for any new phenomenon, so it wouldn't be surprising if a child managed to model the language of its parents just by listening without specific language slots ready to receive types of words
We already know about Brocca's and Wernicke's areas in the brain
I've thought of a way to investigate the Pirahã question of recursivity - Alexa, GTP-3, and a good sat phone, running away snooping 🤪
It‘s all about types. Language is typeborn and typedriven. Types are already present in big apes, some birds, dolphins and even in octopuses. Types provide a digital pattern of information processing totally different from the prior analogue modes . Richard Dawkins postulated memes in 1976, but memes have to have a substrate, an implementation. The downfall of the Journal of Memetics in 2005 was due to this lack of a functional foundation. I think types were at first used internally for fine grained representations and movements. At a certain point they evaded into gestures and sounds. This is the beginning of language. Grammar is in its core a categorical comprehension and expression. Language consists of monoids, monads, semigroups, covariant and contravariant functors etc. So it can be described by the terms and laws of category theory. I think, it makes sense to divide language into a pure internal subsystem, that’s highly composable and insures the fluent production of language and a monadic system, that makes language expressable, for instance in speech, body movement or literature. In contrast to the language of the honey bee, the human language can be easily transformed into infinite subsets of Domain Specific Languages (DSLs). If we want to get any further in language analysis it’s inevitable to take care of the non-communicative foundation of language. Take this two videos to reason about the problem. They show emerging expressiveness of a, in my opinion type driven system, but not in terms of a classic language. th-cam.com/video/NPxgRAp0Z_0/w-d-xo.html
th-cam.com/video/Wos8kouz810/w-d-xo.html
I like slides with information