I see myself agreeing with the fact that language is evolved, since it is clear that all the organs responsible for language have evolved from the great apes for us to produce language. From the brain to our vocal chords since neither of these are suited for language purposes in great apes. I believe that yes there is a part of UG that is innate but that there is also a part of grammar that doesn’t come to us unless we learn through trial and error. I agree with the fact that we could feel that something is wrong with “I yesterday eat an apple” even if we still understand it as “I ate an apple yesterday” but sometimes even people who have gone to school and learned the “rules” of grammar make mistakes, not much of a difference between the 2 examples since you still understand what they are trying convey but one is a little bit better then the other. This also made me think about the difference between us and other primates especially the great apes who have been able to learn around 500-1000 symbols and type them to convey what they meant, I read somewhere that they could even point to what they wanted. This made me think about symbol grounding and how these apes are able to conduct a symbol manipulation of language. Obviously this isn’t all there is to language and it is the way we manipulate these symbols, grammar, that made our language what it is. Moreover as to what made our language evolve symbol grounding isn’t the answer as I agree with the fact that language evolve for social purposes.
On Harnad. L'univers de Chomsky. 2014 : ‘’L’intuition de Chomsky concernant la relation (ou l’interface) entre le sens et la syntaxe est que la langue est pour tout être humain la condition de possibilité de la pensée elle-même et que la possibilité de penser comme telle vient avec certaines contraintes structurelles platoniciennes qui, à leur tour, donnent lieu aux règles de la grammaire universelle. La structure de la grammaire universelle est liée par la structure de la pensée.’’
I am not a Platonist (at least not on this aspect) so I can’t agree with this. I think language is a creation and that thought precedes it, although they necessarily evolve in a close dialectical relationship. But I do agree with the fact that UG may be bound by natural laws, physical laws of the universe, that we have grasped, as a species, through our contact with the exterior world. So we could say that since the physical universe is the condition of possibility of UG, UG cannot violate and has been shaped by the laws of the natural universe. So that violating UG would be like violating the laws of the universe… So actually I guess I’m positioning myself somewhere between the Platonist and the evolutionary view.
‘’Les affirmations non conformes à la grammaire universelle sont soit des pensées mal exprimées, soit des pensées impensables.’’
Ok so I fell into the unthinkability trap. What does it mean to say that sentences that are unthinkable are the ones that violate UG (not some parameter variation)? Or that UG-violating sentences are incorrectly formed or unthinkable? If they are unthinkable how can they be spoken? I am confused, I understand it as saying: everything respects UG; if it doesn’t, it is wrong and when corrected respects UG; all the rest doesn’t –nothing else does- exist. Like, from all utterances that we can think of, those that don’t respect UG are wrong and can be corrected, and there are some other things that exist but that we can’t think of because they don’t respect UG, thus don’t respect the laws of nature (or a subset of them that shapes UG)…
“[Brown and Hanlon] looked for a correlation, but failed to find one: parents did not differentially express approval or disapproval to their children contingent on whether the parents did not differentially express approval or disapproval to their children contingent on whether the child’s prior utterance was well-formed or not (approval depends, instead, on whether the child’s utterance was true.” It seems that the correlational studies of Brown and Hanlon merely test whether parents actually use negative evidence as often as they do positive evidence, and in which instant each form of evidence is used. Does this not reflect the parent’s behavior as opposed to the manner in which the child causally uses each form of evidence? Pinker’s evidence for the independence of negative evidence in eliciting UG is only indirect, pointing more towards the statistical behavior of the parents as opposed to the child. How can we draw concrete conclusions about externalized UG if we are only testing the statistical probabilities of certain inputs? I imagine that more isolated experiments are required to prove the eventually expression of UG without the internalization of negative evidence. An experiment (that is highly unethical, I might add) in which a child is reared in an environment with absolutely no negative evidence will show whether the child is still capable of externalizing UG compliant language. This would eliminate the statistical uncertainty of whether the parents provide larger, or smaller amounts of negative evidence, and in what context it is used.
More so, an even better experiment, would be to provide children with all inputs that do not align with positive evidence. If the child still produces output that corresponds with the UG compliant positive evidence (without ever internalizing it) it would show that language acquisition depends on innate mechanisms only. The child would produce positive output without ever being shown that the way they are speaking is the correct way.
I really liked this paper but I was particularly puzzled by one section; the section that claims children require both positive and negative evidence to acquire their first language. I agree with the necessity of positive evidence because the input that children hear is mostly grammatical. So the grammatical input in the environment is reinforced the more they hear it. What’s troubling to me is the necessity for negative evidence. Negative evidence is the evidence available to the language learner about what is not grammatical (i.e. When a parent corrects a child’s speech, the child has access to negative evidence). But there is no negative evidence in the environment. There are two explanations to support the latter.
First, when a parent attempts to correct the child when he/she produces an ungrammatical sentence, the child ignores the parent and continues to produce the ungrammatical sentences. Second, the parents themselves don’t know the underlying rules, so how could they possibly teach them to the children?
Pinker then goes on to assert that this negative evidence is established by a mental mechanism. That is, there is an underlying mental process that allows children to access negative evidence, and disregard ungrammatical utterances. This may in fact be true. But how can we make such a bold (and almost radical) claim if there is no evidence? Why can’t we conclude instead that children learn language from the positive evidence in their environment, which in turn, rules out the probability of ungrammatical sentences being true?
All in all, I don’t see how we can claim so assertively that there is an underlying mechanism providing the negative evidence to first language learners, when there is clear evidence for positive evidence that is abundant in the environment.
''If children had to learn all the combinations separately, they would need to listen to about 140 million different sentences''
This article by Pinker makes one point; in studying the acquisition of language, it becomes evident that a general human logic cannot be responsible. Without too much direct engagement with Chomsky's principle of Universal Grammar and the poverty of stimulus argument, Pinker makes it clear that it would take something of the order of a a supercomputer (and even then who knows) to acquire language at the rate and with the proficiency and fluency that a human child does. That leaves us with two possible conclusions. One is that children are unimaginable supercomputers who can logic their way into understanding something as complex as languag. The other is that the brain, through evolutionary crystallization, has structures, modules, however you want to call them, which powerfully predispose children (and moreover humans) to the acquisition of language. The former option seems totally implausible, for while children are masters of language acquisition, they are total failures in other things which a general logic, powerful enough to parse out the complexities of language, would easily solve. The second makes the most sense to me. The way I imagine this is that humans come with a set of assumptions which are unbeknownst to the human. It is our mode of being, which we inherit evolutionarily, and which grounds all the experiences we will have. It is our evolutionary narrative, our history as a species, and truly as life. This is our nature (Sidenote: I use nature here to mean naturalized, not as an inherent property of god or the order of the world but instead meaning something we have no access to change or be conscious of - this nature though itself has been produced by a set of unnatural, or incidental occurrences. think of the spandrels from Pinker and Bloom’s Natural Language article) As we go through life, we build other types of knowledge on top of this set of assumptions. We thus come with a set of assumptions pertaining to communication and language, without which language would be impossible to acquire given the other constraints of our nature i.e. our life span, our logic capacities and speeds, etc. I am looking forward to reading critiques of UG because it seems to me undeniable and irrefutable at this moment. How else could language exist without a huge evolutionary handicap? Many point to the impossibility (to imagine) how language or UG evolved in such a short time evolutionarily. That doesn't refute the fact that there is no other way in which language could arise, but instead shows us that evolutionary theory might need some re-examining.
Another post in response to Harnad's article '''Why and How the Problem of the Evolution of Universal Grammar (UG) is Hard.'':
''The hard problem arises because UG has no apparent adaptive advantages. For although a professional grammarian's lifetime is long enough to work out most of UG's rules explicitly by trial and error induction, it turns out that (with the possible exception of a few small portions of UG) no logical or practical advantage has yet been discerned that favors what UG allows over what it disallows, or over an altogether different set of grammatical rules (perhaps even a much simpler and learnable set). The absence of a biological advantage for UG is an even greater handicap than the poverty of the stimulus. It means that even with all of evolutionary time at its disposal, there is no ordinary evolutionary explanation for how or why UG would have been selected (if its basis is the usual genetic variation, selectively propagated through the survival/reproduction advantages it confers on its bearers).''
I think one of two things could explain this. One is that evolution does not create the most biologically advantageous instance that is possible over infinite time (for even evolution has a poverty of stimulus, though much less poor than a human lifespan), but instead supports instances which are more advantageous than others at any given point in time. Thus UG is advantageous in its ability to give humans language, and its specificity is an accident of history. It is possible and probable that UG could and will change through time to a simpler or more efficient set of rules for communication given the environment. Is it not enough that UG allows for language to explain its tendency to evolve, and that any non-advantageous aspects are forgivable given evolution's own poverty of stimulus (read: the finitiude within which evolution works)?
The other is that UG truly does have biological advantages (or is restricted by certain biological truths) in being conditioned by the way we have evolved in the past, and that we just can't understand or have access to these advantages (or agreement with historical constraints/ realities) because of our position in time. We don't have access to our evolutionary history thus UG seems weird and arbitrary. UG has to be in conjunction with a huge set of evolutionary realities pertaining to our bodies and minds - UG is historically constituted. Thus it would be specific, both logical in some ways and totally incidental in others due to its accordance with a history we can never fully know.
I see myself agreeing with the fact that language is evolved, since it is clear that all the organs responsible for language have evolved from the great apes for us to produce language. From the brain to our vocal chords since neither of these are suited for language purposes in great apes.
ReplyDeleteI believe that yes there is a part of UG that is innate but that there is also a part of grammar that doesn’t come to us unless we learn through trial and error. I agree with the fact that we could feel that something is wrong with “I yesterday eat an apple” even if we still understand it as “I ate an apple yesterday” but sometimes even people who have gone to school and learned the “rules” of grammar make mistakes, not much of a difference between the 2 examples since you still understand what they are trying convey but one is a little bit better then the other.
This also made me think about the difference between us and other primates especially the great apes who have been able to learn around 500-1000 symbols and type them to convey what they meant, I read somewhere that they could even point to what they wanted. This made me think about symbol grounding and how these apes are able to conduct a symbol manipulation of language. Obviously this isn’t all there is to language and it is the way we manipulate these symbols, grammar, that made our language what it is. Moreover as to what made our language evolve symbol grounding isn’t the answer as I agree with the fact that language evolve for social purposes.
(I posted on the thread for last week)
DeleteOn Harnad. L'univers de Chomsky. 2014 :
ReplyDelete‘’L’intuition de Chomsky concernant la relation (ou l’interface) entre le sens et la syntaxe est que la langue est pour tout être humain la condition de possibilité de la pensée elle-même et que la possibilité de penser comme telle vient avec certaines contraintes structurelles platoniciennes qui, à leur tour, donnent lieu aux règles de la grammaire universelle. La structure de la grammaire universelle est liée par la structure de la pensée.’’
I am not a Platonist (at least not on this aspect) so I can’t agree with this. I think language is a creation and that thought precedes it, although they necessarily evolve in a close dialectical relationship. But I do agree with the fact that UG may be bound by natural laws, physical laws of the universe, that we have grasped, as a species, through our contact with the exterior world. So we could say that since the physical universe is the condition of possibility of UG, UG cannot violate and has been shaped by the laws of the natural universe. So that violating UG would be like violating the laws of the universe… So actually I guess I’m positioning myself somewhere between the Platonist and the evolutionary view.
‘’Les affirmations non conformes à la grammaire universelle sont soit des pensées mal exprimées, soit des pensées impensables.’’
Ok so I fell into the unthinkability trap. What does it mean to say that sentences that are unthinkable are the ones that violate UG (not some parameter variation)? Or that UG-violating sentences are incorrectly formed or unthinkable? If they are unthinkable how can they be spoken? I am confused, I understand it as saying: everything respects UG; if it doesn’t, it is wrong and when corrected respects UG; all the rest doesn’t –nothing else does- exist. Like, from all utterances that we can think of, those that don’t respect UG are wrong and can be corrected, and there are some other things that exist but that we can’t think of because they don’t respect UG, thus don’t respect the laws of nature (or a subset of them that shapes UG)…
“[Brown and Hanlon] looked for a correlation, but failed to find one: parents did not differentially express approval or disapproval to their children contingent on whether the parents did not differentially express approval or disapproval to their children contingent on whether the child’s prior utterance was well-formed or not (approval depends, instead, on whether the child’s utterance was true.”
ReplyDeleteIt seems that the correlational studies of Brown and Hanlon merely test whether parents actually use negative evidence as often as they do positive evidence, and in which instant each form of evidence is used. Does this not reflect the parent’s behavior as opposed to the manner in which the child causally uses each form of evidence? Pinker’s evidence for the independence of negative evidence in eliciting UG is only indirect, pointing more towards the statistical behavior of the parents as opposed to the child. How can we draw concrete conclusions about externalized UG if we are only testing the statistical probabilities of certain inputs? I imagine that more isolated experiments are required to prove the eventually expression of UG without the internalization of negative evidence. An experiment (that is highly unethical, I might add) in which a child is reared in an environment with absolutely no negative evidence will show whether the child is still capable of externalizing UG compliant language. This would eliminate the statistical uncertainty of whether the parents provide larger, or smaller amounts of negative evidence, and in what context it is used.
More so, an even better experiment, would be to provide children with all inputs that do not align with positive evidence. If the child still produces output that corresponds with the UG compliant positive evidence (without ever internalizing it) it would show that language acquisition depends on innate mechanisms only. The child would produce positive output without ever being shown that the way they are speaking is the correct way.
DeleteI really liked this paper but I was particularly puzzled by one section; the section that claims children require both positive and negative evidence to acquire their first language. I agree with the necessity of positive evidence because the input that children hear is mostly grammatical. So the grammatical input in the environment is reinforced the more they hear it. What’s troubling to me is the necessity for negative evidence. Negative evidence is the evidence available to the language learner about what is not grammatical (i.e. When a parent corrects a child’s speech, the child has access to negative evidence). But there is no negative evidence in the environment. There are two explanations to support the latter.
ReplyDeleteFirst, when a parent attempts to correct the child when he/she produces an ungrammatical sentence, the child ignores the parent and continues to produce the ungrammatical sentences. Second, the parents themselves don’t know the underlying rules, so how could they possibly teach them to the children?
Pinker then goes on to assert that this negative evidence is established by a mental mechanism. That is, there is an underlying mental process that allows children to access negative evidence, and disregard ungrammatical utterances. This may in fact be true. But how can we make such a bold (and almost radical) claim if there is no evidence? Why can’t we conclude instead that children learn language from the positive evidence in their environment, which in turn, rules out the probability of ungrammatical sentences being true?
All in all, I don’t see how we can claim so assertively that there is an underlying mechanism providing the negative evidence to first language learners, when there is clear evidence for positive evidence that is abundant in the environment.
''If children had to learn all the combinations separately, they would need to listen to about 140 million different sentences''
ReplyDeleteThis article by Pinker makes one point; in studying the acquisition of language, it becomes evident that a general human logic cannot be responsible. Without too much direct engagement with Chomsky's principle of Universal Grammar and the poverty of stimulus argument, Pinker makes it clear that it would take something of the order of a a supercomputer (and even then who knows) to acquire language at the rate and with the proficiency and fluency that a human child does. That leaves us with two possible conclusions. One is that children are unimaginable supercomputers who can logic their way into understanding something as complex as languag. The other is that the brain, through evolutionary crystallization, has structures, modules, however you want to call them, which powerfully predispose children (and moreover humans) to the acquisition of language. The former option seems totally implausible, for while children are masters of language acquisition, they are total failures in other things which a general logic, powerful enough to parse out the complexities of language, would easily solve. The second makes the most sense to me. The way I imagine this is that humans come with a set of assumptions which are unbeknownst to the human. It is our mode of being, which we inherit evolutionarily, and which grounds all the experiences we will have. It is our evolutionary narrative, our history as a species, and truly as life. This is our nature (Sidenote: I use nature here to mean naturalized, not as an inherent property of god or the order of the world but instead meaning something we have no access to change or be conscious of - this nature though itself has been produced by a set of unnatural, or incidental occurrences. think of the spandrels from Pinker and Bloom’s Natural Language article) As we go through life, we build other types of knowledge on top of this set of assumptions. We thus come with a set of assumptions pertaining to communication and language, without which language would be impossible to acquire given the other constraints of our nature i.e. our life span, our logic capacities and speeds, etc.
I am looking forward to reading critiques of UG because it seems to me undeniable and irrefutable at this moment. How else could language exist without a huge evolutionary handicap? Many point to the impossibility (to imagine) how language or UG evolved in such a short time evolutionarily. That doesn't refute the fact that there is no other way in which language could arise, but instead shows us that evolutionary theory might need some re-examining.
Another post in response to Harnad's article '''Why and How the Problem of the Evolution of Universal Grammar (UG) is Hard.'':
ReplyDelete''The hard problem arises because UG has no apparent adaptive advantages. For although a professional grammarian's lifetime is long enough to work out most of UG's rules explicitly by trial and error induction, it turns out that (with the possible exception of a few small portions of UG) no logical or practical advantage has yet been discerned that favors what UG allows over what it disallows, or over an altogether different set of grammatical rules (perhaps even a much simpler and learnable set). The absence of a biological advantage for UG is an even greater handicap than the poverty of the stimulus. It means that even with all of evolutionary time at its disposal, there is no ordinary evolutionary explanation for how or why UG would have been selected (if its basis is the usual genetic variation, selectively propagated through the survival/reproduction advantages it confers on its bearers).''
I think one of two things could explain this. One is that evolution does not create the most biologically advantageous instance that is possible over infinite time (for even evolution has a poverty of stimulus, though much less poor than a human lifespan), but instead supports instances which are more advantageous than others at any given point in time. Thus UG is advantageous in its ability to give humans language, and its specificity is an accident of history. It is possible and probable that UG could and will change through time to a simpler or more efficient set of rules for communication given the environment. Is it not enough that UG allows for language to explain its tendency to evolve, and that any non-advantageous aspects are forgivable given evolution's own poverty of stimulus (read: the finitiude within which evolution works)?
The other is that UG truly does have biological advantages (or is restricted by certain biological truths) in being conditioned by the way we have evolved in the past, and that we just can't understand or have access to these advantages (or agreement with historical constraints/ realities) because of our position in time. We don't have access to our evolutionary history thus UG seems weird and arbitrary. UG has to be in conjunction with a huge set of evolutionary realities pertaining to our bodies and minds - UG is historically constituted. Thus it would be specific, both logical in some ways and totally incidental in others due to its accordance with a history we can never fully know.