Many people have argued that the evolution of the human language faculty cannot be explained by Darwinian natural selection. Chomsky and Gould have suggested that language may have evolved as the by‐product of selection for other abilities or as a consequence of as‐yet unknown laws of growth and form. Others have argued that a biological specialization for grammar is incompatible with every tenet of Darwinian theory ‐‐ that it shows no genetic variation, could not exist in any intermediate forms, confers no selective advantage, and would require more evolutionary time and genomic space than is available. We examine these arguments and show that they depend on inaccurate assumptions about biology or language or both. Evolutionary theory offers clear criteria for when a trait should be attributed to natural selection: complex design for some function, and the absence of alternative processes capable of explaining such complexity. Human language meets this criterion: grammar is a complex mechanism tailored to the transmission of propositional structures through a serial interface. Autonomous and arbitrary grammatical phenomena have been offered as counterexamples to the position that language is an adaptation, but this reasoning is unsound: communication protocols depend on arbitrary conventions that are adaptive as long as they are shared. Consequently, language acquisition in the child should systematically differ from language evolution in the species and attempts to analogize them are misleading. Reviewing other arguments and data, we conclude that there is every reason to believe that a specialization for grammar evolved by a conventional neo‐Darwinian process.
Saturday 11 January 2014
8a. Pinker, S. & Bloom, P. (1990). Natural language and natural selection
Pinker, S. & Bloom, P. (1990). Natural language and naturalselection. Behavioral and Brain Sciences13(4): 707-784.
Many people have argued that the evolution of the human language faculty cannot be explained by Darwinian natural selection. Chomsky and Gould have suggested that language may have evolved as the by‐product of selection for other abilities or as a consequence of as‐yet unknown laws of growth and form. Others have argued that a biological specialization for grammar is incompatible with every tenet of Darwinian theory ‐‐ that it shows no genetic variation, could not exist in any intermediate forms, confers no selective advantage, and would require more evolutionary time and genomic space than is available. We examine these arguments and show that they depend on inaccurate assumptions about biology or language or both. Evolutionary theory offers clear criteria for when a trait should be attributed to natural selection: complex design for some function, and the absence of alternative processes capable of explaining such complexity. Human language meets this criterion: grammar is a complex mechanism tailored to the transmission of propositional structures through a serial interface. Autonomous and arbitrary grammatical phenomena have been offered as counterexamples to the position that language is an adaptation, but this reasoning is unsound: communication protocols depend on arbitrary conventions that are adaptive as long as they are shared. Consequently, language acquisition in the child should systematically differ from language evolution in the species and attempts to analogize them are misleading. Reviewing other arguments and data, we conclude that there is every reason to believe that a specialization for grammar evolved by a conventional neo‐Darwinian process.
Many people have argued that the evolution of the human language faculty cannot be explained by Darwinian natural selection. Chomsky and Gould have suggested that language may have evolved as the by‐product of selection for other abilities or as a consequence of as‐yet unknown laws of growth and form. Others have argued that a biological specialization for grammar is incompatible with every tenet of Darwinian theory ‐‐ that it shows no genetic variation, could not exist in any intermediate forms, confers no selective advantage, and would require more evolutionary time and genomic space than is available. We examine these arguments and show that they depend on inaccurate assumptions about biology or language or both. Evolutionary theory offers clear criteria for when a trait should be attributed to natural selection: complex design for some function, and the absence of alternative processes capable of explaining such complexity. Human language meets this criterion: grammar is a complex mechanism tailored to the transmission of propositional structures through a serial interface. Autonomous and arbitrary grammatical phenomena have been offered as counterexamples to the position that language is an adaptation, but this reasoning is unsound: communication protocols depend on arbitrary conventions that are adaptive as long as they are shared. Consequently, language acquisition in the child should systematically differ from language evolution in the species and attempts to analogize them are misleading. Reviewing other arguments and data, we conclude that there is every reason to believe that a specialization for grammar evolved by a conventional neo‐Darwinian process.
Subscribe to:
Post Comments (Atom)
“When one looks at more abstract linguistic analyses, the underlying unity of natural languages is even more apparent. Chomsky has quipped that anything you find in one language can also be found in every other language, perhaps at a more abstract level of representation, and this claim can be justified without resorting to Procrustean measures.”
ReplyDeleteDoes this suggest that innate grammar preceded the development of these various languages? I keep thinking of observations on Nicaraguan Sign Language, a relatively new language that has been under observation since its ‘creation’. Studies have demonstrated that each new generation of deaf signers that enters the community modify and change the language, and with each new generation of modifications, NSL is becoming more and more similar to the grammar of mature sign languages in terms of the ways signs are used to express different ideas. For example, the younger signers were able to learn the basic NSL established by the previous generation, and modify it by adding spatial specifications in their sign, so that signing in a particular area of space in front of them would signify one object, while another spatial location would identify another object, and so forth, whereas the older signers were not found to have developed this more complex and specific way of communicating. Although it doesn’t tell us about how the language faculty evolved in humans, it does demonstrate that UG has an immense effect on creation of languages, and that we are, in a way, wired to communicate in through this intricate system.
The article highlights that human language has evolved because of natural selection, partially based on the fact that ‘the only explanation for the origin of organs with complex design is the process of natural selection’, but there is no ‘organ’ for innate grammar? Or, perhaps, are they referring to the physical specificities of our vocal tract that allows us to produce the sounds we are able to produce? If this latter idea is the case, I don’t think it’s necessarily true – several structures (ex. tongue, epiglottis, etc) have survival functions as well.
1. If there is an inborn universal grammar (UG), then all members of our species have it now. When it first appeared in the evolution of language is another question. And in general explaining the evolution of UG (not the evolution of language, but the evolution of UG) is a special problem -- and one that Pinker & Bloom fail to mention or face: they only mention the "easy" problems of language evolution.
Delete2. Sign language are languages, like all others, so they are all compliant with UG. But not all grammatical rules are part of UG. Some are arbitrary conventions, just like vocabulary, that are not shared with other language, and they are learned rather than innate. That is probably what is still developing in NSL. (It's not clear which commonalities among sign languages are because of UG and which are because of the necessary demands of visible manual signing in space.)
3. The brain is the organ for cognition, including language (and UG), whether the languages are vocal, written, or gestural. If UG exists and is innate, its brain mechanism would be the organ.)
First off, this is the very first paper on linguistics I've read - if I'm missing something by all means let me know.
ReplyDeleteOne of the things which caught my attention was when Pinker discussed inflectional parings.
ice-infested (OK); rats-infested (bad)
men-bashing (OK); guys-bashing (bad)
teethmarks (OK); clawsmarks (bad)
feet-warmer (OK); hand-warmer (bad)
purple people-eater (OK); purple babies-eater (bad)
Those made sense to me, with the exception of hand-warmer. Hand-warmer seems valid, and in google-ing it, I am able to find eponymous products. Did Pinker mean to state "hands-warmer", as this is incorrect, and in line with the other "bad" examples in that it is plural.
Dealing with the concept of universal grammar, which we've discussed in class and is discussed in the article, I've read the following article in the past which critiques it. As I mentioned, I don't know very much about linguistics, and am uncertain if this is a legitimate counter-argument.
http://www.newyorker.com/magazine/2007/04/16/the-interpreter-2
It discusses Dan Everett, a linguist who claims to have discovered a tribe where principles of Universal Grammar are disobeyed. For example, recursion does not occur in the same way. Maybe someone with a linguistics background could comment on the validity of the argument? Even Pinker himself has commented on these claims.
Below is an relevant excerpt from the article:
"Everett, who this past fall became the chairman of the Department of Languages, Literature, and Cultures at Illinois State University, has been publishing academic books and papers on the Pirahã (pronounced pee-da-HAN) for more than twenty-five years. But his work remained relatively obscure until early in 2005, when he posted on his Web site an article titled “Cultural Constraints on Grammar and Cognition in Pirahã,” which was published that fall in the journal Cultural Anthropology. The article described the extreme simplicity of the tribe’s living conditions and culture. The Pirahã, Everett wrote, have no numbers, no fixed color terms, no perfect tense, no deep memory, no tradition of art or drawing, and no words for “all,” “each,” “every,” “most,” or “few”—terms of quantification believed by some linguists to be among the common building blocks of human cognition. Everett’s most explosive claim, however, was that Pirahã displays no evidence of recursion, a linguistic operation that consists of inserting one phrase inside another of the same type, as when a speaker combines discrete thoughts (“the man is walking down the street,” “the man is wearing a top hat”) into a single sentence (“The man who is wearing a top hat is walking down the street”). Noam Chomsky, the influential linguistic theorist, has recently revised his theory of universal grammar, arguing that recursion is the cornerstone of all languages, and is possible because of a uniquely human cognitive ability.
Steven Pinker, the Harvard cognitive scientist, calls Everett’s paper “a bomb thrown into the party.” For months, it was the subject of passionate debate on social-science blogs and Listservs. Everett, once a devotee of Chomskyan linguistics, insists not only that Pirahã is a “severe counterexample” to the theory of universal grammar but also that it is not an isolated case. “I think one of the reasons that we haven’t found other groups like this,” Everett said, “is because we’ve been told, basically, that it’s not possible.” Some scholars were taken aback by Everett’s depiction of the Pirahã as a people of seemingly unparalleled linguistic and cultural primitivism. “I have to wonder whether he’s some Borgesian fantasist, or some Margaret Mead being stitched up by the locals,” one reader wrote in an e-mail to the editors of a popular linguistics blog."
[Continued]
DeleteFinally, the article discussed child-language learning, as distinct from adult language learning. I understand there are substantial differences in neurological development at the age, but I'm not convinced this is absolutely related to language learning differences. Yes, adults never tend to master languages the same way that children do, but there is a plethora of confounds related to this difference. Children, when learning languages essentially have full time exposure to the target language, whereas most adults learning a new language have very limited exposure. The time which may be devoted to language learning differs substantially; children will almost always have more hours in a day, and years to learn it, which is not the case with adults. Furthermore the differences in responsibilities are substantial.
I am willing to believe a child's language learning is different than an adults. I just am skeptical of the sufficiency of the evidence presented in its favour, and that the confounds I brought up (as well as others) that are not addressed.
Andras,
DeletePirahã language is spoken by humans. Humans are bound biology which is messy and not entirely predictable.
Supposedly Pirahã language lacks recursion. Even if it's the case, we still have to explain the-over-five-thousand other languages that do use recursion.
Just adding my two cents as a linguistics student:
DeleteI think that UG and strong pro-Chomsky views on language can be a lot to wrap one's head around, especially if it's your first real brush with linguistics as a field. Linguists and psychologists don't always see eye to eye when it comes to language and that makes it all the more complicated.
If you're feeling skeptical about acquisition, you might want to look into the Variational Model theories (like Yang, 2002) of language ac. They're a little more intuitive and represent one of the many alternative theories to strong Chomsky UG that (in my opinion) are often overlooked.
Chomsky and Gould “have repeatedly suggested that language may not be the product of natural selection, but a side effect of other evolutionary forces such as an increase in overall brain size and constraints of as-yet unknown laws of structure and growth”.
ReplyDeleteWorking backwards, from my 8b skywriting, I feel as though this explanation causes issues to my thinking (it’s one of these questions: which came first, the chicken or the egg?). So Pinker and Bloom (1990) are arguing that language is the product of natural selection, which I agree with their argument/s. But [and I could be making a detrimental mistake here] what if the quote is actually trying to get at the question of ‘motivation’ [motivation of creating a proposition/symbol system] ((the question of motivation is brought up in 8b)- and of course by assuming this as a potential answer we are left with the question of why did the brain size increase). The brain difference between the chip’s brain size and human brain size are quite drastically different, but again my problem is did the start-up of language cause for this change? Or was it the change in size which started-up the language indirectly through motivation?
The authors do a good job at arguing that evolution and language as an adaptation are better able at explaining certain mysteries that we had with other theoretical hypotheses. And they do a good job arguing that brain size alone can not account for language, they give an analogy: adding more neuronal circuits to a brain will not emerge any new abilities, rather it is the formation of very well designed and structured neural circuits that will allow for new abilities (pg.35). Furthermore, they explain that the size of the brain increased do to the social aspect of how language allowed for cheating, and the brain developed as a means to detect for this (pg.46-7). However, even with this explanation I still feel like this philosophical conundrum still hasn’t been answered for me. This question is positing on the distinction between software and hardware, is motivation something that would be attributed on a level of the software (therefore the brain size shouldn’t matter), or hardware (therefore the bran size could be a culprit of accountability)? And am I making the big mistake of seeing this as a dualist?
Hi Demi,
DeleteCoincidentally enough, my two points about this article are in the same line of your argument:
1. I was, too, surprise of the Pinker and Bloom arguing about the motivational component of the emergence of language. In the sense that because we were apparently more social animals than non-human primates, we had a greater motivation to name and describe everything. However, I don't think that motivation and the process of natural selection counter-argue with each other (If I understand correctly your inquiry).
I have seen in other classes arguments that motivations could partly have a genetic basis (e.g. Ericsson argues that talents is 0% genes and 100% work, but he does conceives that genes may influence the motivation to do the necessary work). Thus, via natural selection, those who appeared more motivated to name and describe everything could have more chance at reproduction (the smooth-talking argument that the Prof described last week) and at surviving overall (by learning new categories orally).
2. I also like the analogy with neural circuits, and I have had a similar parallel in an other class regarding talent. Taking Ericsson again, you aren't born with talent; you are born with oligodendrocytes and schwann cells that produces myelin sheets around your neurons. The more you use a particular network, the more efficacy it gains over time. Thus, as Pinker and Bloom would have put it; it not that you'll have the network for all natural language build in innately. Rather, humans would have simply been quicker and quicker at forming these networks.
(Although this is a strech; I am not sure whether we have significantly more myelin that non-human primates)
So the neural network formation could be a hardware-thing (thus dependent on brain size), but I don't see why motivation should be pitched in the hardware/software distinction. Motivation is very complex, and it should probably have a genetic basis that's also influenced by non-biological factors (e.g. motivation to increase smooth talking could have also emerged from seeing other people being successful at seducing with language)
I actually just realized that it was in Harnard's article that the motivation argument was discussed; not from Pinker and Bloom
DeleteIt's both! Professor Harnad kind of threw up his hands wondering why other Hominidae don't speak. It seems like they can figure out that one item is a conjuct of two others, so why not propose the case?
DeletePinker and Bloom answer that monkeys are just not as motivated to mind-read.
“within a group of interdependent, cooperating individuals, the states of other individuals are among the most significant things in the world worth knowing about. Thus communication of knowledge and internal states is useful to creatures who have a lot to say and who are on speaking terms.”
I'm not sure if this is much of an explanation though. Kind of just re-wording the probelm. I remember these videos of Washoe and Nim Chimsky, chimps trained to speak sign language. They hardly said anything. They were capable of answering questions, for treats. But spontaneous utterances? Only "hungry," and "bannana."
“[T]here are clear criteria for when selectionist and nonselectionist accounts should be invoked to explain some biological structure: complex design to carry out some reproductively significant function, versus the existence of a specific physical, developmental or random process capable of explaining the structure’s existence.”
ReplyDeleteThis is for sure an important criterion to keep in mind. However, its efficiency entirely depends on our ability to tell on which side of this ‘versus’ falls a given function. The problem is the heavily nonlinear dynamics (a system of infinitely many variables where a small change in one variable can result in a complete reorganization) behind the coming about of biological forms. Our linear stories can hardly deal with the emergent properties of highly complex systems (qualitative changes at the global level that cannot be reduced to the local), so we are likely to have a bias towards seeing design where there is none.
“[C]onvergent evolution, resemblance to man-made artefacts, and direct assessments of engineering efficiency are good sources of evidence for adaptation.”
This is apparently a decisive argument in the paper, but I don’t see why we should buy it. Our artefacts often use biological functions and entities as models so it is not true that they can serve as an independent criterion. Worse: the camera is useless without an eye, therefore its existence presupposes that of the eye. “It is impossible to make sense of the structure of the eye without noting that it appears as if it was designed for the purpose of seeing...” But does it make sense to speak of the purpose of seeing if there are no eyes? The authors try to break the circularity by saying that “biologists posit far fewer functions than there are biological systems”, but that’s just abstraction. My seeing and the bee’s seeing are the same function in some analogical sense, but a different function in the sense that it can see things I cannot see.
“But is there any reason to believe that there are as yet undiscovered theorems of physics that can account for the intricate design of natural language? Of course human brains obey the laws of physics, and always did, but that does not mean that their specific structure can be explained by such laws.”
I think that Chomsky has in mind, not physical laws like gravity, but laws such as the aforementioned emergent properties of complex systems. An ant colony shows behaviour many would call cognition yet nobody would presume that the rules are coded anywhere, let alone passed on genetically.
“Arguments that language is designed for communication of propositional structures are far from logical truths” (section 3.2)
ReplyDeleteHere, Pinker refutes the idea that language merely exists as a symbol system for the representation of internal human reasoning. He cites a few reasons for this.
1. Natural languages are ineffective for this. Languages are not always the best way to spell out reason, as they demand certain grammatical structures and are tailored to clear communication in a way that makes them more complex.
2. Language is not designed to merely express thought, but rather in order to communicate. This piggy-backs off of the last point.
3. Communities share common languages - people do not have individualized languages. Language thus functions to connect communities around shared thoughts.
I don’t see how these reasons prove his point. This section is focused on showing that the evolution of language is not a just-so story. To do this, he relies on several arguments that seem to me to be merely based on intuition (especially his first 2 arguments). He says that language is useful for communication in specific and complex way, and therefore language must be “designed” and not a mere by-product of another evolved structure. The argument I delineate above has the most substantial explanation and examples, but I still struggle to grasp what he is getting at. Pinker shows that language is not the best medium for precise logical reasoning and that language’s primary function is to allow the sharing of thoughts and information. He further admits that language is not the best way to express emotions or display identity. But at the same time, he wants us to see that language is so complex and precise that it must have evolved or been “designed” for its specific function. Towards the end of the section he says that languages displays, “non-accidental parallels that are reminiscent of the talk of diaphrams and lenses when applied to cameras and eyes”. These are examples of how items that are truly designed for a particular function mirror the natural products of evolution. All that this example shows me is that human designers like to take advantage of nature’s mechanisms rather than reinventing the wheel. Likewise, programming languages are structured in parallel with human language because it makes them more useful and understandable to us, not because they are the only feasible systems for representing our thoughts. I am very confused by what Pinker is arguing here, as his arguments do not all seem to point in the same direction and do not all seem substantially supported.
“Natural selection…is the only physical process in which the criterion of being good at seeing can play a causal role. As such it is the only process that can lead organisms along the path in the astronomically vast space of possible bodies leading from a body with no eye to a body with a functioning eye” (section 2.2)
ReplyDeleteThis statement attributes causal power to the ability to see. Pinker explains that good vision caused an increase in reproductive success for organisms with this ability, and thus they were able to survive and spread their genes to offspring and perpetuate the genetic underpinnings of sight. Can we think about sight and language in this exact same way? How is language genetically heritable? It seems like the biological structures that allow us to speak are heritable, but language itself is merely enabled by the existence of these things like vocal chords and proper neural circuits. Language itself is the cause of higher reproductive success in organisms with vocal cords and proper neural circuits, but does this mean that language itself was selected for? Or is language a by-product of selection for biological structures? It depends on how we speak about evolution - either with genes as the object of selection or higher order functions, like language or vision itself, as the object of selection.
In section 5, Pinker addresses the fact that genes are the substrate of evolution. He says, language must be more than useful, it must have a history of genetic variation between organisms that resulted in different language abilities. He provides evidence that there is variation in the genetic underpinnings of organisms’ language abilities and gives further proof that other requirements of evolution could have existed (i.e. shared genetic mutations allowing communication even at the first stages of language, section 5.2). This genetic evolution seems like enough for noisemaking, but organized language systems sophisticated enough for communication seem designed in a way that only human intention, not the blind selection of evolution could have created. Pinker says, “Pidgins, contact languages, Basic English, and the language of children, immigrants, tourists, aphasics, telegrams, and headlines provide ample proof that there is a vast continuum of viable communicative systems displaying a continuous gradation of efficiency and expressive power. This is exactly what natural selection requires”. This spectrum of communication methods is not underpinned by genetic variation. These are examples of different human-designed outcomes all taking advantage of the same biological capacity for speech and communication. Likely, I’m just missing the subtleties of Pinker’s explanation of exactly what mechanism of inheritence underpins language, but I’m still trying to wrap my head around it.
Weintraub,
DeleteSo you're sitting there wondering whether or not language is in fact a just-so story, a spandrel?
You could tell this story: in as much as we have articulators, and the trigeminal nerve along with four other nerves innervating facial muscle, we have language. We never needed, but now that we have it, there's no going back. Like pants. Who could imagine a backwords world where people don't where pants?!
For one thing, and you know this, we don't need to make noises to make language, evidenced by sign language. Similarly, without the ability to make noises, we will eventually wind up making language, as evidenced by la Idioma de Señas de Nicaragua. What could we lack that would prevent a Homo sapiens spapiens from doing language?
Tear down the arch, the spandrel disapears. What do we remove that would allow us to say, "ah, all along we only had language because we had ...?" Lesion stuides primarily look at the arresting of speech, which means the articulators, not language per se.
Alex, I like how you raised lesion studies. When we perturb the brain's ability to communicate verbally or physically, we see that the lesioned individual becomes disadvantaged.
DeleteYet do they become disadvantaged because everyone around them relies on a language in which they can no participate? It still does not really address the question of whether the purpose language serves is the reason it evolved. If we become accustomed to using a specific tool in a particular way, then regardless of whether that tool evolved for the use we bestowed upon it, we will suffer because we have lost the tool.
I guess what I am now getting at is multiple uses/purposes/explanations?
It may be the case that, once we know what every protein coded by every gene in every strand of DNA does, we will find that the tenets of Universal Grammar are indeed innate and hardcoded. However, under the continued rationale that ‘if something can be externally coded rather than innate, then it is likely to be externally coded’, I would still like to posit a critique of the innateness of language.
ReplyDeleteThat humans have *something* innate which makes them, and them alone, the animal species able to engage in this sort of communication is obvious to everyone. The disagreements begin in concluding what is innate. The minimalist interpretation is that language is a purely emergent property from having a larynx suitable for speech sounds and sufficient intelligence to generate a code. The maximally innate view is Universal Grammar. I wouldn’t go so far as to call myself a minimalist, but I believe that Pinker glosses over the in-tenability of Universal Grammar far too quickly in his work.
“While one might justifiably argue that an entire system of grammar must evolve in a gradual continuous sequence, that does not mean that every aspect of every rule must evolve in a gradual continuous sequence.” So we allow for partial grammars that are useful at every stage, but incorporate more and more complexities as time goes on. But the fact remains – why on earth must we appeal to an innate mechanism for this? Pinker admits the importance of being able to distinguish “whether a far‐off region is reached by taking the trail that is in front of the large tree or the trail that the large tree is in front of. It makes a difference whether that region has animals that you can eat or animals that can eat you.” And so on. But languages would be a lot more similar than they actually are if Universal Grammar were coded internally via genetics than externally via every culture’s independent, ingenious ways of solving the quandary of information transmission and distinctions such as those enumerated by Pinker.
Sentences that are ambiguous in some languages have no ambiguity in others, but no language has ambiguity in every sentence; some languages take fewer words to say some things, and more to say others, but no language requires a minimum of ten words to form a grammatical phrase; some languages focus on word order to convey meaning, some languages focus on aspect/tense derivations, but no spoken language relies on how many fingers the speaker is holding against her temple at the time of uttering each word to determine the word’s lexical role. So we divide “all possible languages” into “human” and “non-human” grammars, and the alternative explanations are – there is a universal grammar that evolved into being, or that things were done certain ways because they make sense given the other physical and perceptual constraints of humans. I strongly believe the latter. Some itineration’s of Universal Grammar attempt to explain every human language in terms of parameters and principles, when in reality, the space denoted by Universal Grammar is so vast that it ends up being a post-hoc explanation of grammars as it encounters them. Universal Grammar is useful as a window into what humans can and cannot do with language – but why rely on genetics?
Incidentally, I have my doubts as to the "unlearnability" of Universal Grammar. We can't explain exactly how it is we learn to ride a bike - nor can we make a convincing automaton that is able to ride a bike - but we don't need anything that drives us innately to be able to bike beyond our general abilities of balance, navigation, and instruction-following. Just because something is subconsciously or non-declaratively learned does not make it innate.
DeleteIn this piece, Pinker and Bloom argue that “language has been shaped by natural selection.” This means that it “belongs more to the study of human biology than human culture.” Basically, language is an adaptation and humans who had this trait were more likely to survive and reproduce, passing this trait on to their kin. Language must, in Pinker and Bloom’s account, have some basis in genetics.
ReplyDeletePinker and Bloom’s argument is twofold: first, they show that language belongs to the category of things that can be explained by evolutionary biology and then they refute major arguments that have been levelled against the claim that the origin of language can be explained by natural selection. I will talk about the first part of their argument in this blog.
Pinker and Bloom explain that not all biological structures can be explained by natural selection. For example, natural selection isn’t particularly helpful in explaining biological structures that aren’t involved in any function or behaviour, such as why there is a v-shape between your fingers. It also isn’t that helpful in explaining why a structure that is used for something really complex (ex: a bird’s wings, engineered for flying) is used for something fairly simple too (blocking the sun).
But natural selection is super power for explaining things that “complex design to carry out some reproductively significant function.” Like your eye. I mean, seeing is definitely helpful if you are trying to survive and reproduce! And the eye is an incredibly complicated structure, “[i]t is impossible to make sense of the structure of the eye without noting that it appears as if designed for the purpose of seeing.”
So the question: is language in the category of biological structures that can be explained by natural selection?
Pinker and Bloom clearly think language is in the “complex design to carry out some reproductively significant function category” (read: the natural selection category). First, grammar is really complex. (Pinker and Bloom get into all sorts of gritty linguistic details here about the building blocks of the “universal grammar” that underlies all languages – but honestly, I was willing to just take their word for it!) Second, language is reproductively significant: we cannot physiologically adapt quickly enough to deal with the environmental changes that may come our way, so it is a good thing we have a way of transmitting information to deal with new changes and threats! As the authors write, “[t]here is an obvious advantage in being able to acquire such information second-hand: by tapping into the vast reservoir of knowledge accumulated by some other individual, one can avoid having to duplicate the possibly time-consuming and dangerous trial and error process that won that knowledge.” It is faster and safer for me to learn which mushrooms are dangerous by being told how to spot the poisonous ones than by having to taste them (risking illness or death) all by myself!
(Is this where Harnad et al got their inspiration for their account of the origin of language?)
“Symonsʹ (1979) observation that tribal chiefs are often both gifted orators and highly polygynous is a splendid prod to any imagination that cannot conceive of how linguistic skills could make a Darwinian difference.”
ReplyDeleteThere were so many interesting points in this article that it was difficult to settle on just one or a few topics that I wanted to talk about in my post, but in the end I decided to latch onto this excerpt, because it relates to what I wrote in the skywriting for the second article – I did some armchair philosophizing about the difference between the evolution of language on a geological scale vs. the evolution of linguistic ability on an individual scale. Both the articles seem to be aiming at an explanation of the geological evolution of language, that is, how we came to acquire the ‘language organ’, if such a thing exists. The other type is a linguistic evolution and it is more fine-grained – it relates to how we improve our vocabulary or speaking skills on a personal level. I think (or at least, I convinced myself in the other skywriting) that there is a fundamental difference between these two, so it’s surprising to me that in this article, an observation about one (linguistic skill of tribal chiefs) is equated with the other (evolution of language ability as a whole). I don’t doubt that both types of evolutions are adaptive, but they just take place on such different scales – “linguistic skills” can conceivably be acquired in one’s lifetime with practice; whereas language ability is, as the article argues, something that evolved over thousands of generations. Furthermore, while I buy the arguments that natural selection was the ‘programmer’ that led humans to acquire the language organ, I don’t believe natural selection works in the same way on one’s public speaking skills or vocabulary. Being a good speaker may be adaptive, but it’s not the kind of skill that determines whether you live or you die, and as far as I know it’s not something that can be encoded and passed down in one’s genes. So, given the huge difference between linguistic ability and language ability, is it acceptable to make such a leap of a comparison at all? I think it’s a dangerous move when presenting an argument, and that there is much better support for the Darwinian evolution of language presented in the rest of the paper, but I was wondering what everyone else thought.
The way I understood the excerpt you posted was that either the tribal chief became a good public speaker because his position of leadership demanded him to practice and eventually become better and better (adaptation) or he had a natural knack for powerful speech, which gave him social (and evolutionary) advantage.
DeleteSo in response to your question: "Given the huge difference between linguistic ability and language ability, is it acceptable to make such a leap of a comparison at all?"
No! Public speaking is a skill that some people have a talent for, but that can also be improved.
Speaking are a physiological action of saying words in order to communicate is a basic ability that everyone possesses.
So I agree that it's a leap.
So I looked into the Piraha a bit after reading Andras’ comment to see what Daniel had to say about the lack of recursiveness in their language, and the implications that would have with respect to language generally and universal grammar specifically.
ReplyDeleteIt seems that while the Piraha cannot embed phrases within their syntax, they can nevertheless express recursive statements. For example, instead of saying, “the cat in the hat is mine” (with “in the hat” as the recursive part), they would say, “The cat is in the hat. It is mine”. In this way, the Piraha can satisfy the definition of a language that Dr. Harnad gave in the first lecture: It can verbalize all that is verbalizable.
With that property established it becomes a very tenuous attack on universal grammar to remark on the fact that Piraha language grammar does not allow embedded clauses, given that they can still express the recursive propositions that embedding was supposed to be crucial for in the first place! In other words: the function remains, but the function’s implementation that changes.
I do not have much experience with linguistics but from what I gather universal grammar is just a set of theoretical properties that all languages share, and these properties do not necessarily have to be expressed in the syntax of each individual language. If the ability to express recursive propositions really is at the core of human language capabilities, as Chomsky believes it is, then the question becomes whether or not it was developed through natural selection, saltatory evolution, or as a spandrel of something else. To be honest, It seems to me like it could have been any of them. To even understand, let alone express, recursive propositions, you would need a cognitive mechanism to arrange objects of thought hierarchically, knowing which objects are affected by particular properties of other objects. It is not unfeasible to imagine how understanding what is essentially a logic operation would be useful for all sorts of cognition, so whether it developed as part of language or if part of language’s evolution was to simply recruit this mechanism from other parts of the brain is underdetermined. Either way, even if evolving an understanding recursion was a gradual process, it would have created an explosion of progress in language evolution as soon as it became operational.
http://edge.org/conversation/recursion-and-human-thought
According to the article on the Piraha tribe, recursion is “a linguistic operation that consists of inserting one phrase inside another of the same type, as when a speaker combines discrete thoughts (“the man is walking down the street,” “the man is wearing a top hat”) into a single sentence (“The man who is wearing a top hat is walking down the street”).” From what I understand, they would therefore not replace “the cat in the hat” with “it” in the example you gave. Piraha speech consists of only direct sentences. An example given in the article is the following : “I saw the dog. The dog was at the beach. A snake bit the dog” instead of “I saw the dog that was down by the river get bitten by a snake”. According to Chomsky’s theory of Universal Grammar “recursion is the cornerstone of all languages, and is possible because of a uniquely human cognitive ability.” So even though there we do still have over 5 thousand other languages, as Alex stated in Andras’ comment, the Piraha language does make us question Universal Grammar. I think this is a interesting counterexample to UG, and it does favour Pinker and Bloom opinion that human language evolves by natural selection. This tribe was almost completely isolated from the outside world and their language has evolved into a completely different manner. The Piraha language is based on 8 consonants and 3 vowels and integrates whistling, singing, humming within the language. Very interestingly, they have no numbers and can’t perform tasks that involve more than three things. Instead they use sentences such as “a small size or amount” and “a somewhat larger size or amount. They don’t name colors either; instead they will say “It looks like blood” for the color red.
DeleteAdditionally, the Piraha tribe has shown extreme resistance to other cultures, which is hypothesized to be one of the reasons their language has developed in a completely different manner. However, Chomsky believes that cultures play a little role with the study of language. Pinker and Bloom seem to disagree with Chomsky on this point as well. In Section 5.3.3. and 5.3.4 they show the influence of culture in shaping language. “In sum, primitive humans lived in a world in which language was woven into the intrigues of politics, economics, technology, family, sex, and friendship and that played key roles in individual reproductive success.” Politics, economics, technology family etc. play an immense role in forming culture. The Piraha tribe seems to be another example of the influence of culture. Being isolated from the outside world, their culture has evolved in an entirely different way and so has their language.
Let's distinguish between what the Piraha people have done so far, and can or could do in future, with the brains they already have. And let's also distinguish what you can say in Piraha, so far, and what you can or could say in Piraha in future, if you needed to. Numbers have not yet been lexicalized and recursive embedding has not yet been syntacticized. But they could be, even tomorrow, if you like.
DeleteChomsky is talking about a grammar that is universal to all humans (UG) that allows them to learn and speak any language at all. As we learn in week 9, this "learning" is unusual, in that UG itself is not (and cannot be) learned by the language-learning child because its database (what it hears, says, and gets corrected for) is not sufficient to learn the rules of UG. Linguists can learn them, based on decades of collaboration, guided by their own linguistic intuitions about what can and cannot be said. But the language-learning child can't. It just hears UG-compliant sentences. And it only utters UG-compliant sentences. So it's never corrected. It never gets to compare the UG-compliant and the non-UG-compliant utterances, the way the linguists do. That's the "poverty of the stimulus." So UG is innate.
But we are talking about UG without even knowing what it is: At least the UG linguists know what rules are at issue here. We don't.
Recursion is not UG. Recursion is just one of the properties of UG -- if it's used. All natural languages comply with UG, but they differ in a few (learnable) parameters. Some drop pronouns -- Vengo instead of I'm coming -- no need for the I. That's a parameter, Subject/Verb/Object vs S/O/V is another parameter. I'm not sure there's a "recursion-on" parameter (though I suspect there isn't, and that Paraha is already recursive as used now); maybe.
Now the isolation is over, Paraha is probably already going the same way as all other languages...
“All we have argued is that human language, like other specialized biological systems, evolved by natural selection.”
ReplyDeleteIt seems to me that if universal grammar didn’t evolve be natural selection, there isn’t a viable alternative route by which humans could have come to have these abilities. Although, Pinker and Bloom do a good job neutralizing arguments against the possibility that language evolved via natural selection, the issue for me has never been one of determining whether language evolved. I’m more concerned with elucidating the specific reproductive advantage that the building blocks to universal grammar provided their possessors, and ascertaining what the first building blocks actually were.
Pinker and Bloom suggest that human ancestors fell into a lifestyle where cooperation was key if one wanted food, safety and reproductive opportunities. This means that it would be advantageous to reap the benefits of this cooperative society while finding subtle ways to avoid paying the costs. (where the costs are doing the work required to find and catch the food to be shared, building the shelter etc.) It would also be beneficial for non-cheaters to notice cheaters because when there aren’t any cheaters the work is divided up most evenly. So the pressure to detect subtle signs of cheating and the pressure to find less detectable ways of cheating resulted in a cognitive arms-race according to Pinker and Bloom. They point out that language would be advantageous in this arms-race as it would allow one to verbally frame offers in a way that make them sound beneficial to the receiver when they are really more beneficial to the giver. This makes sense, and it also makes sense that language would be advantageous in cooperative societies. However, Pinker and Bloom are unable to suggest when the building blocks to language first emerged or what they might have been. Based on this article, it is unclear whether the cooperative societies were the product of the emergence of the initial building blocks of language or if cooperative societies emerged first and the building blocks of language emerged second because cooperative societies made them advantageous.
“The first recognizably distinct mental system that constituted an antecedent to modern human language may have appeared in a species that diverged from the chimp‐human common ancestor, such as Australopithecus Afarensis or any of the subsequent hominid groups that led to our species. Moreover chimpanzees themselves are not generalized common ancestors but presumably have done some evolving of their own since the split.”
As a side note, I think that the part of the paper quoted above, where Pinker and Bloom explain that there might not be any living primates with homologues of human language is really important. The assumption that the fact that humans and apes have a common ancestor from several million years ago means that the gestures and vocalizations made by apes must have been the precursors to universal grammar in humans is not valid. I’m not saying that looking to apes cannot be useful, I just think that it needs to be taken with a grain of salt. Ape gestures and vocalizations may be no closer to universal grammar than dog gestures and vocalizations. The other paper for this week points out that apes seem to have the intelligence necessary to learn language but not the motivation. Looking into what kind of alterations in social structure and environment could foster motivational changes may be a good pathway take primate language research on.
I've been thinking a lot on your last point too. What came first? The tools for the communication of language or the cognitive Universal Grammar? The tools for the communication of language seems to be exaptations--limbs are used for picking up things, the mouth for the consumption of food. However, I find it hard to see UG as a exaptation, unless one was to conclude that language is just like any other learned skill.
DeleteI don’t think UG is an exaptation but other cognitive processes are. UG’s principles result from cognitive capacities we had before language emerged and rely on pre-existing structures used to perform other tasks.
Delete“Our conclusion is based on two facts that we would think would be entirely uncontroversial: language shows signs of complex design for the communication of propositional structures, and the only explanation for the origin of organs with complex design is the process of natural selection” (Pinker & Bloom, 1990).
ReplyDeleteThe conclusion that Pinker and Bloom are referring to is that human language evolved by natural selection, and their paper makes a thorough argument to support this. This is quite a controversial argument, as most psychologists and cognitive scientists would argue that it is doubtful that Darwin’s principle of natural selection could explain the evolution of a biological specialization for grammar, but I think that Pinker and Bloom make a convincing argument in their paper. I think it is interesting that Pinker and Bloom focus on the specific organs used in speech to make their case, as the opposing theories on the evolution of natural language tend to ignore this. After all, to create speech, one must have the physical organs that are necessary to produce speech, so it cannot be a solely cognitive process. That being said, the organs may be necessary, but are in no way sufficient to produce speech, so there must have been a cognitive mechanism which perhaps organized and allowed for natural language to develop. I do think that the cognitive organizational effects of language must have developed before the physical organs used to speech. Many animals can communicate, although they do not seem to have a sophisticated language like ours; even the involvement of the tongue in speech production appears to be a human phenomenon. All in all, I think that Pinker and Bloom have developed an intuitive argument to support the idea that language evolved through the Darwinian concept of natural selection. I also think that is was rather brave of them to make such a claim given that the majority of the scientific community would disagree.
“It is certainly true that natural selection cannot explain all aspects of the evolution of language”
ReplyDeleteIn this article by Pinker and Bloom there were many interesting things to talk about, I didn’t really know the origin of language could be so controversial, our use of it is so natural (for lack of a better word) it is just “there;” that it is hard to think at some point in the history of our species it wasn´t there. I want to focus here on one of the strongest evidence I believe they presented for a theory of natural selection in the evolution of language; particularly in point 4.2 on “Constrains in Possible Forms.” Pinker and Bloom argue against the idea that “the mind is an all-purpose learning device” and Gopnik’s developmental dysphasia example is strong evidence against the idea that language is just a consequence of how human brains developed, in this case “the sufferers lack control of morphological features such as number, gender, tense, and case” while being intellectually normal. As they point out “mere largeness of brain is neither a necessary nor sufficient condition for language” nonetheless I wonder why they did not use examples of Broca or Wernicke’s aphasia. People with Broca’s aphasia (in very simplistic terms) speaking is damaged while comprehension is preserved and in Wernicke’s aphasia the inverse (again I’m simplifying a lot), but doesn’t this speak to some kind of modularity that could have evolved as an organ? I am against trying to localize “things” in the brain, in the end it always goes back to networks (as I pointed out in an earlier skywriting) but I wonder why Pinker and Bloom decided to leave out these disorders that result from brain damage. I also think it is important that they admit it cannot be natural selection all the way down and that there are “aspects of the system that can only be explained by historical, developmental and random processes” if not, it would just amount to another “just-so story.” Language is result of “evolutionary pressures” and as it is discussed in the following reading a result of motivation. This type of motivation relates, in my view, a lot to what we were discussing in the last seminar: what determines fitness? Professor Harnad mentioned a lot “talking the talk,” so it’s easy to see how language must have benefitted fitness: men trying to court women with language. Considering, though, that it was only “talking the talk” what got language started seems erroneous on its own, like any other of these theories, it would also be a “just-so story.” I think most of the “just-so stories” could have contributed to the emergence of language, that “talking the talk” proved to also be beneficial when a mother was trying to calm-down her new born baby by singing, these could all be instances of motivation, and acted in conjunction to be preserved in our brain.
“Seidenberg and Petitto (Seidenberg, 1986; Seidenberg and Petitto, 1979, 1987) have reviewed the evidence of the signing abilities of apes and concluded that they show no significant resemblance to human language or to the process of acquiring it….There just arenʹt any living creatures with homologues of human language” (p 48).
ReplyDeleteIn the last section of their article, Pinker and Bloom distinguish language as distinctly human and argue against the claim that we share language capacities with other animals (such as chimpanzees). Despite the great amount of genetic overlap between humans and species such as chimpanzees, language remains a distinct human faculty. I personally believe this to be true; just because a chimpanzee is able to use the ASL sign for water when he/she is thirsty, it does not mean that the chimpanzee has language capacities.
However, some still believe that other animals do have the capacity for language, even if the way that the animals communicate is through other modalities like hand gestures. Pinker and Bloom do identify language as distinctly human, but do so to highlight the fact that the mental capacity for language evolved after divergence the chimp-human split. I was still curious as to how exactly human language differed from how chimpanzees use sign language.
So, I did some more research looking at the productions of famous signing chimpanzees such as Washoe and Nim Chimpsky and reading Chomsky’s interview with Matt Aames Cucchiaro titled “On the Myth of Ape Language”. One must remember to distinguish lexicon from grammar and realize that they are both necessary to produce and comprehend language. Just because a chimpanzee has learned to associate a certain action with a certain item or has learned to imitate a human, it does not the animal is capable of language. The chimpanzee is does follow the grammatical rules. For example, Nim Chimpsky produced strings such as “apple me eat” and “grape eat Nim” through ASL. While Nim has learned to associate certain actions with objects and actions, it is clear that his free word order style of language in addition to the absence of case markings violates certain universal principles of human language. It seems that the use (and the innateness) of grammar and formal rules is what distinguishes human language from chimpanzee signing.
Notes:
Nim Chimpsky productions found here: http://www.sciencemag.org/content/206/4421/891
Chomsky Interview found here: http://www.chomsky.info/interviews/2007----.htm
I definitely agree with what you said confusing lexicon with language (that chimpanzees cannot learn language). I think that the key lies in the complexity of grammatical rules which we somehow seem to understand and know, without explicitly being taught the grammatical rules for (the idea of universal grammar). I guess the thing that I was most confused about was, then, how did universal grammar evolve, if it does exist? I would like to add though, that even though Nim Chimpsky seemed to produce the right actions, that there is no way to know whether he really understands what those hand actions mean, or whether that series of actions will result in him being rewarded with an apple or some other kind of food reward. I would say that there is little distinguishing what Nim Chimpsky did from behaviourist learning. Another interesting note about ASL, though: children that are born deaf have grammatical nuances in their signing that are not as fluid or present if one becomes deaf later in life and acquires ASL!
DeleteOverall, I would have to agree with Pinker and Bloom that language is unique to humans and that at some point it did not exist (at least not in the form it does today). Given the complexity and all the nuances of language, it seems that Pinker and Bloom make a lot of sense when they say that: since language is so complex in its realization (i.e. design), it doesn't make sense that it spontaneously appeared, but that it evolved over time throughout natural selection (especially given the clear advantage that conferring information through language gives over learning through trial and error).
In this paper Natural Language and Natural Selection, Pinker and Bloom defend their position that "language has been shaped by natural selection" (page 3). Later in this paper, they emphasize that if we want to account for adaptive complexity of language, we can only use natural selection. They also say that "language evolution and language acquisition not only can differ but also they must differ" (page 31). In order to demonstrate their points, they make an analogy between language and spandrels, saying that language is like a spandrel derived from some other adaptations. Also, they mention Chomsky's Universal Grammar in their paper.
ReplyDeleteI pretty much can understand what Pinker and Bloom are trying to say. However, what they actually mean by "language" is not quite clear to me. Language is quite enormous and multifacet, and I believe it is not tenable to say language is shaped by natural selection without indicating what exactly language is that they are talking about. From my previous classes and what I have read and known so far, there are not really many disputes about the theory that Chomsky's Universal Grammar is innate (maybe there are some disputes, but it seems that most people and my professors agree that it is innate), and I believe it is innate too because infants do have some judgments towards language or are born with some sort of language ability. In terms of Universal Grammar in language, I do agree with Pinker and Bloom that it is shaped by natural selection. However, just as they claim that language evolution and language acquisition are two separate things, I also believe Universal Grammar cannot be all the grammar we use today and that there is another type of grammar that is totally separate from Universal Grammar. It is not hard to understand because we learn grammar such as tense, plural, conjugations of verbs in our English classes. They also mention in their paper that we can understand She comed although She came is the right way to say and that children who say She comed at first will change to She came afterwards. Clearly it is a different type of grammar from Universal Grammar because children do not know come should become came instead of comed in past tense unless someone told them or they hear people say it. This type of grammar, I believe, thus is not shaped by natural selection and cannot be explained by evolutionary psychology; it is learned by people after they are born, and I think it is the part Pinker and Bloom fail to account for or neglect in their paper.
With that being said, I do appreciate most of what they say in their paper (especially the analogy between language and spandrels although I believe language is far more complicated than a spandrel in terms of what defines them and what their functions are). I always hold the opinion that language is shaped by an interaction of natural selection and learning. Or to make myself clearer, people do not learn everything about language after they are born, we are born with some knowledge about language already. It is a combination of both instead of solely one component. However, I could not say what Pinker and Bloom say is wrong, because natural selection does play a big role in language evolution.
Also, I want to criticize one of the many ways they use to defend themselves because I found it kind of funny when I read the paper. Sometimes they seem to use the rule that "since there is no enough evidence, you cannot say it is wrong". Well, I could not say this is wrong either, but as a scientific and academic paper, it would be arbitrary to say your argument is not wrong or right because there is no enough evidence (and this is only my personal opinion).
There is not much I can say about this article. I whole-heartly agree with everything written. How does your iPhone understand what you want to say better than you do? That’s not fair, but why do we understand it to be this way? Sometimes when one person has felt robbed of their own voice, it is extremely difficult to maintain the structure and “rules” of ordinary grammar. In this way, it may be easier to comprehend oneself and their surroundings by using devices such as computers, cameras, etc. to enhance already-formed neurophysiological processes. We need not forget that the brain is plastic and it is important to understand that our own plasticity is something that can be used to improve not only one being but many.
ReplyDeleteI love what you said about my iPhone knowing better what I want to say than I do and how we can feel robbed of our own voice when under stress or heightened emotions.
DeleteThe last bit about relying on devices, computers etc. for language/communication reminded me of the extended mind thesis. If you already know about it, then you can stop reading, but here is a link that summarizes the idea (that our minds are not demarcated by our brains and that we extend our cognition into the environment all the time):
http://consc.net/papers/extended.html
ReplyDeleteIn his article Pinker shows that arbitrary language constraints can actually be adaptive solutions by referring to the example of subjacency. This constraint prevents movement that take a constituent out of certain phrases from occurring and it exists to prevent a speaker from formulating a sentence with a too complex structure that a listener could not understand.
But then, why then isn’t the canonical word order kept? For example, in English questions are formed with movements but it is not the case in Chinese. It would therefore be possible to have questions keeping the canonical word order and it would then be easier to understand which is the thematic role of a word used in a sentence.
But I have another question: assuming that there is a universal underlying structure of language, does this underlying structure correspond to a step in the evolution of language? For example, some priming effect studies have shown that fillers (moved elements) are reactivated in the reader at a gap site (the place where the filler should be if the sentence had surfaced with its canonical word order). Could it be anevidence that at some point in evolution, moved elements surfaced in their original position?
Pinker and Bloom's argument that language evolution was mediated through natural selection hinges on one main point: that language is complex and specialized, not unlike the vertebrate eye, and therefore must have been "designed" through a process of natural selection. This is not as obvious in language processing as in the eye, because the structures are not specifically designed for transduction of physical stimulus, but instead purposed to the rather vague notion of "propositional representation". But if not from natural selection, then where does this propositional representation come from? The alternative argument seems to be for a general increase in cognitive capability, or some other scaling up of our ability to represent knowledge combined with our physical ability to communicate. What Pinker and Bloom seem to not quite arrive at is that this "scaling up" is incredibly vague, and could be mediated through language itself. That there exists rules for a Universal Grammar is not necessarily an infallible proof for grammar being hard wired, but rather could be evidence of knowledge representation being hard wired - the increase in our ability to represent information about the world inside of our heads is the evolution of language. This seems to fit nicely with the fact that we often think and problem solve in verbalizable structure. If language were just a system for sharing what we already are able to hold in our minds, why would we think in words?
ReplyDelete"...language is complex and specialized, not unlike the vertebrate eye, and therefore must have been "designed" through a process of natural selection. This is not as obvious in language processing as in the eye, because the structures are not specifically designed for transduction of physical stimulus."
DeleteThis is simultaneously very interesting but highly frustrating to think about. The eye is so much simpler with rods and cones we can identify and test but how did Wernicke's and Broca's areas develop to be language specific parts of the brain? Did we develop them as our brains changed over time? Were they laying dormant? The brain and language development can only give correlational data but nothing causal. It seems rather narrow so say language has evolved just as a matter of natural selection.
ReplyDeleteI can't help but sit back in awe at the universality of language proficiency. Any sort of genetically related language acquisiton mechanism must have been come-to when the human races was in one concentrated area. The argument that asks "how a hypothetical 'beneficial' grammatical mutation could really have benefited its possesor, given that none of the person's less evolved compatriots could have understood him or her," reminds us that, according to what we udnerstand of natural selection and mutations, there was theoretically at one point less than a mere handful of people who had this mutation! Here we are, typing, speaking in class (sometimes even speaking French or Hungarian), using what came of that first mutation along with the 7 billion other language users who have essentially the exact same starting equipment. It is astounding that we are all capable of aquiring any language as infants so long as we are exposed to them. It is an incredible feat of nature that is probably rarely recognized for its absurdly complicated make-up and its universality. In my mind, it is analogous to transporting an person from the the 15th century to any modern city and watching them marvel at the wondrously complex and ubiquitous smartphone. So many things must go right in order to communicate via language, and in almost every case it works out just fine. Pretty impressive if you ask me.
“It would be natural, then, to expect everyone to agree that human language is the
ReplyDeleteproduct of Darwinian natural selection. The only successful account of the origin
of complex biological structure is the theory of natural selection, the view that
the differential reproductive success associated with heritable variation is the
primary organizing force in the evolution of organisms (Darwin, 1859; see
Bendall, 1983 for a contemporary perspective). But surprisingly, this conclusion
is contentious. Noam Chomsky, the worldʹs best‐known linguist, and Stephen
Jay Gould, the worldʹs best‐known evolutionary theorist, have repeatedly
suggested that language may not be the product of natural selection, but a side
effect of other evolutionary forces such as an increase in overall brain size and
constraints of as‐yet unknown laws of structure and growth.”
While Pinker and Bloom posit that it is “surprising” for people to disagree on the fact that human language is a product of Darwinian natural selection, I don’t find this surprising at all, even within the context of things that are acceptedly evolutionarily. Regardless of the philosophical debate, very few things in the discussion of natural selection are ever not disagreed with and debated, even within the field of evolutionary biology. In Pinker and Bloom’s comparison of language evolution with the evolution of other capacities, they seem to disregard the complexity of the issue of evolution altogether. Sure, very few educated, intelligent people deny evolution, but this doesn’t mean evolutionary theories are always accepted and agreed upon right away as fact. In reality, the way in which things evolved, and the nature of any specific thing’s evolution, is often disagreed upon.
Language also differs greatly from capacities such as echolocation and stereopsis, which Pinker and Bloom compare it to, because of how quickly and consistently it is developing and changing (or “evolving”). Language is much more difficult to explain evolutionarily than other inherent abilities because it doesn’t exhibit the very slow pattern of change that other faculties do. It also seems short-sighted to so quickly deny the impact of “other evolutionary forces” when discussing the evolution of language.
While I don’t strongly disagree with Pinker and Bloom, I am not sure if I am completely convinced, either. I can buy their big-picture argument about language and natural selection—that is believable in a general “makes sense” kind of way—but it still seems like this paper is missing a crucial point. It is also difficult for me to think of the evolution of *language* as separate/different from the evolution of *UG* because it seems that if UG was truly universal, it would have been present from the start.
ReplyDeleteAdditionally, part of my skepticism stems from he emphasis on syntax alone—Pinker and Bloom largely ignore other parts of language that could be related to evolution, like phonetics (i.e. when did we acquire the capacity to control our articulators? How is our articulatory system uniquely human?). Also, the way that tradeoffs for utility in a language are described does not convince me, as a reader, that evolution happens on the level of individual grammars. Why can’t UG have evolved to handle differed morphological systems, instead of Turkish and Serbian being completely different? Pinker addresses this issue on a per-language basis, it seems, and skirts around answering when and how UG came into the human cognition—I think tackling those questions would offer more insight and bolster this argument. Pinker seems to have a lot of 'whys' and relatively few 'hows'.
This comment has been removed by the author.
ReplyDeleteIn the portion of this paper about phyletic continuity (which I find extremely interesting) Pinker and Bloom argue against Bates, Greenfield and Lieberman's argument that there must be a biological ability shared by humans and related species (chimpanzees) that precedes and is the basis for our development of language. They argued that this 'continuity' must be present if linguistic ability evolved through natural selection. Research done by Seidenberg and Petitto show that there is no significant similarity between the signing abilities of apes and human language and that language and non-linguistic gestures are completely dissociable. This would demonstrate (assuming the above theory is correct) that the 'discontinuity' of human linguistic abilities with those of other primates proves that language could not have evolved by natural selection. The authors rebut this conclusion beautifully by explaining that chimpanzees and humans have perhaps evolved separately after splitting from a common ancestor. They explain that universal grammar may lie within the 1% small, yet significant, genetic gap between us. The significance of this gap is made more significant by the possibility that one specific base pair on a gene might completely change the functioning of the entire gene (which is extremely interesting to me.)
ReplyDeleteI don't understand, however, why the continuity between language and nonlinguistic neural mechanisms (which Lieberman believed existed between Syntax and brain mechanisms devoted to motor control, due to their heirarchical organization) would be relevant to the genetic continuity/discontinuity between us and similar species. I don't understand why the similarity of a linguistic ability to a neural construct would be an important factor to consider. Is it because both species' brains are composed of similar motor areas, and if syntax is necessarily a result of 'motor programs' then we could validly conclude that continuity exists?
I don’t really have any substantial critique of Pinker/Bloom’s paper. I really do think they’re on point in defending the claim that our language faculties must have been sculpted through natural selection. The only leaning I have that might deviate from them (Pinker at least since I’ve read his books) is that I think a more substantial part of the seeming deep structure of language can be attributed to the way the world is (the correlational structure of the world and what meanings are more easily acquired through induction and how some substantive universals might be products this induction process), rather than how our brains are initially programmed. Nonetheless, this disagreement is extremely tangential and based on a lot of possibly unwarranted assumptions.
ReplyDeleteIn terms of relevance to our central topic (understanding, explaining and/or creating cognition), the paper itself is a little tangential. We’re talking about the WHYs of cognition rather than the HOWs most of the time. The HOW portion, I think, can be summarized as simply saying, UG must be accounted for in terms of genetics somehow since we presume it was naturally selected.
So, let me see what I can add to this discussion.
“Evolutionary theory appears to have very little to say about speciation, or about any kind of innovation. It can explain how you get a different distribution of qualities that are already present, but it does not say much about how new qualities can emerge.” (Chomsky)
This passage honestly left me confused. How does evolutionary theory not explain speciation? Time and evolution and distinct genetic pools explains speciation rather neatly. New qualities? Mutation and then natural selection does that job… or is there something I’m missing. This seems pretty straight forward. Maybe someone can clue me in as to how a giant like Chomsky could be making this claim.
“A skeptic might accuse the proponent of circularity, asking why a lump of clay should not be considered well‐designed to fulfill the function of taking up exactly the region of space that it in fact takes up.” (Pinker/Bloom)
This is how I do away with it. Take the hypothetical lump of clay, scramble its contents/molecules and it’ll still do the job finely of “taking up exactly the region of space that it in fact takes up”. The function of “taking up exactly the region of space that it in fact takes up” isn’t nearly complex enough to demand any complex design work. Everything we know that exists performs that function. As far as I’m aware only non-scrambled humans use language.
Many have critiqued the paper for its focus on natural selection when making sense of evolved features, and I think that could be fair if the authors were actually talking about just features of any organism. I’m a little put off by this reaction seeing as its pretty clear the authors are talking about UG and/or language capacity which are extremely complex and functional features. Mutations and drift occur in evolution, and mutations are necessary for evolution to occur, but natural selection is really where the actions at so-to-speak. Natural selection is what does the work. Maybe I’m just some neo-darwinism fanboy, but I can’t help think “What’s the alternative?” when someone questions natural selection’s role in designing complex organisms.
“Moreover, any code as complex and precise as a grammar for a natural language will not wear its protocol on its sleeve. No mortal computer user can induce an entire communications protocol or programming language from examples; thatʹs why we have manuals. This is because any particular instance of the use of such a protocol is a unique event accompanied by a huge set of idiosyncratic circumstances, some relevant to how the code must be used, most irrelevant, and there is no way of deciding which is which.” (Pinker/Bloom, p.31)
This passage was enlightening and depressing all at once.
“This is consistent with the speculation that the multiplicity of human languages is in part a consequence of learning mechanisms existing prior to (or at least independent of) the mechanisms specifically dedicated to language. Such learning devices may have been the sections of the ladder that evolution had no need to kick away.” (Pinker/Bloom, p.42)
DeleteI think their emphasis on the Baldwin Effect (as well as prof. Harnad’s) to be essential to the story. It really helps understand how things could get off the ground by positing learning as a precursor to selection for an inherent capacity.
“...what Dawkins (1986) calls the Argument from Personal Incredulity. The argument draws on peopleʹs poor intuitive grasp of probabilistic processes, especially those that operate over the immensities of time available for evolution.” (Pinker/Bloom, p.43)
I enthusiastically agree. Not only that, but (I know we’ve moved past this… I haven’t!) I’d argue the intuition being pumped by Searle through his CRA is just that.
Okay, sorry. Back to the topic at hand! I’d also add that people are incredulous due to Orgel’s 2nd rule which says “Evolution is cleverer than you are.” Essentially, trial and error sometimes can find solutions to a problem thinking through a problem cannot, especially when there’s millions upon millions of trials occurring every minute over stupidly large time scales.
“An oft‐noted special feature of humans is that such knowledge can accumulate across generations.”
I’ve always found two facts fascinating that contribute to this idea. One of these is the fact that civilization most likely started so late after we had the genes and thus phenotype needed to contribute to building a civilization. Secondly, the reality of the Flynn Effect makes me believe cultural/memetic evolution is more important than we normally make it out to be.
While reading this paper, I found myself asking about the function of language. Associating a name or label to an object/thing, knowing what the object/thing is, and communicating that knowledge in a way that is interpretable and agreed among others. This includes understanding the relationship between two things as well as knowing how to manipulate them.
ReplyDeleteTherefore, language may be interpreted as a form of symbol grounding. However, in this case, it would be symbol grounding for a community (as opposed to an individual).
When we agree on something through language, that something grounds the word that we assign to it. This article cleared up some issues I had with the definition of categorization. I initially thought that categorization was assigning a name to a particular thing/object (i.e. using language to describe a particular thing in the universe). For example, when we spoke about the categorization of a zebra, I associated this with learning the word “zebra” and using that as a label (name) for an animal that has black and white stripes. But now I understand that symbol grounding and categorization for the individual is irrelevant to language.
The question then becomes, what is the point of symbol grounding or categorization in a society?
Pinker and Bloom argue for the social value of language. That is, the ability to talk about “time, possession, beliefs, desires, tendencies, obligations, truth, probability, hypotheticals, and counterfactuals.”
This conclusion assumes that humans primarily live socially and that language evolved as a simple tool to “promote” a social life. This idea puzzles me. Wouldn’t this just be another “just so” story of evolution? Given that we live in groups now, we come up with explanations as to how this is possible. Or rather, would it be that, because language exists, and because we live in groups, we use one as an explanation for the existence of the other?
What bothers me most about the arguments proposed by Pinker and Bloom is that they consider language as an organ: "language shows signs of complex design for the communication of propositional structure, and the only explanation for the origin of organs with complex design is the process of natural selection," (50). I find the assumption that language is a equivalent to an organ to be absurd.
ReplyDeleteIf we define an organ as a part of an organism that is a distinct structure and has a specific vital function, then I am even less inclined to agree. Language involves not only the brain but the mouth, throat, larynx, nasal cavity, etc. All of these things need to work together for the production of speech - a vital part of language and communication.
I also find it hard to wrap my head around how this "language organ" developed gradually. And what happened first? Did we have thoughts and no ability to communicate them? Did everything kind of come together over time? What would have been the point of that? And why are the authors comparing the evolution of language to the evolution of an eye? I find myself also wondering that can't a single genetic mutation produce startling and significant changes in the structure of an organism?
Also, is it enough that language passes the tests of "when it is appropriate to invoke natural selection as an explanation for the evolution of some trait," (5)? Perhaps it passes, but is it the whole story?
I find myself having a lot of trouble with the proposed idea that natural selection is the answer to the development of language in humans. I would be interested in other potential theories as proposed by Chomsky and Gould.
“Pidgins, contact languages, Basic English, and the language of children, immigrants, tourists, aphasics, telegrams, and headlines provide ample proof that there is a vast continuum of viable communicative systems displaying a continuous gradation of efficiency and expressive power. This is exactly what the theory of natural selection requires.”
ReplyDeleteIn my opinion, this passage best sums up Pinker’s argumentation for the natural selection of language, in which he expresses a multitude of evolutionary properties associated with natural language, yet unconvincingly provides a clear evolutionary mechanism by which it comes to fruition (specifically through the attainment of universal grammar). Pinker makes clear that the adaptive features of language are analogically related to the evolutionary properties that govern evolutionary mechanism that have been rigorously studied. Pinker notes the advantages of arbitrariness in language design, expressing that its design is a perfect trade of between the demands of the speaker and hearer for the most stable and effective form of communication. In this way, all speakers and listeners are forced to learn language in the same way as all the other individuals trying to learn that language. This mirrors the evolutionarily stable strategies that occur amongst various other forms of selective ecological dynamics.
Furthermore, Pinker seems to provide rather strong evidence that language is not merely a spandrel (an epiphenomenon), by indicating that as children we are taught a language by interacting with it, rather than learning its rules. As consequence, the rules of language, which we intuitively pick up, must be innate, having been “programmed” into us by natural selection. But, it is in this last example that Pinker seems to contradict himself. In this instance he has expressed that variations in natural language, the learned capabilities of children taught by their parents, is not the program, and thus does not undergo selection. It seems like Pinker implies that these variations are really spandrels of the universal language inherent to all of us. So far, then, it is the fundamental basis of universal grammar that is the result of selective pressures.
It is actually in showing the evolution of universal grammar where Pinker seems to lose his credibility. The very nature of a theory, including a theory of language evolution, requires the capability of being falsifiable. I would imagine that proper hypothesis testing in evolutionary sciences, requires a concise outline of the evolutionary mechanism governing the selection of language, and a method to test it. I does not seem clear that Pinker’s explanation of the adaptive selection for universal grammar is testable, nor falsifiable, and makes me wonder, no matter how convincing his argument may sound, that his idea is merely speculative. Of course, speculation may lead to eventual testable evidence, but for now it is just as unprovable as the spandrel-arguments of his rivals. Of course, due to the massive credibility of natural selection, I am inclined to trust his intuition, but what I can’t ignore is that more than just natural selection seems to be required in order to understand cognitive phenomena.
"language shows signs of complex design for the communication of propositional structures, and the only explanation for the origin of organs with complex design is the process of natural selection"
ReplyDeleteI think what this paper managed to demonstrate (aside from its basic conviction quoted above) is that theories of language are often derailed by a misunderstanding of how different theories within various fields of science converge on the topic of language. As the author points out, different approaches to understanding language have stemmed from specific fields of thought and often manage to leave out important knowledge from other fields. I'm not sure that any fully comprehensive knowledge on the evolution or origins of language can be known until scientists have developed a better understanding of the organs/oganisms that give rise to language. I also think part of the problem is that many theories attempt to explain all aspects of language under one umbrella where it might be more relevant to look at different aspects of language separately. Understanding how Universal Grammar might have originated and evolved is probably the most difficult task and it is one that is not much discussed in this paper, however other relevant features can more easily be theorized about.
I don’t have much to say about this article in the way of critique, so I’ll try to summarize my understanding.
ReplyDeleteIt seems that language, whether it was “designed to” or not, plays an important role in our survival as a species. It is a specialized ability requiring a complex design of organs; it is an ability like echolocation, not like writing (in that the complexity of language does not correlate with technological progress, it was not invented by some groups and then spread to others, and that language creates the complex selective advantage that somewhat superficially [from an evolutionary perspective] extends our language ability). Given all this, the origin of language is a problem that falls squarely in the camp of biology, and biology best explains it by natural selection.
The authors of this paper defend a selectionist ‘orthodox synthetic’ view of language evolution, against the Chomsky/Gould view that language has emerged through the brain’s evolution, as a by-product (i.e. innate generative grammar has not evolved by natural selection). Their argument is based on the fact that language fits the criteria for traits acquired through natural-selection: that the complexity and specialization of the system suggest it is a trait that has evolved with a purpose, by natural selection. However a permeable mistake in this text is that they confound language and UG, or language capacity and its product (languages with syntax). This is maybe due to an erroneous reading of the fact that ‘all languages are complex computational systems employing the same basic kinds of rules and representations, with no notable correlation with technological progress: the grammars of industrial societies are no more complex than the grammars of hunter‐gatherers; Modern English is not an advance over Old English.’ This confusion leads them to errors such as mistaking creation of new words for modification of grammatical rules (section 5.2.1).
ReplyDelete‘Finally, there may be direct evidence against the speculation that language is a necessary physical consequence of how human brains can grow. Gopnik (1990a,b) describes a syndrome of developmental dysphasia whose sufferers lack control of morphological features such as number, gender, tense, and case. Otherwise they are intellectually normal. One 10‐year‐old boy earned the top grade in his mathematics class and is a respectable computer programmer. This shows that a human brain lacking components of grammar, perhaps even a brain with the capacity of discrete infinity, is physically and neurodevelopmentally possible.’
Maybe it’s the combination of ‘necessary’ and ‘can’ in the negated proposition of the first sentence that makes me twitch, but it seems obvious to me that if 1) language exists and 2)is supported by the brain, then yes, there is a way in which the brain grows that necessarily entails language ability. This does not exclude the possibility that some brains can grow lacking this ability, it just means that there exist different brain systems that are sufficiently independent, at least initially, in order for them to develop in the absence of the others. Maybe, by appropriately distinguishing language capacity and universal grammar, we could accept that language as a capacity evolved by natural selection, but that UG –or syntax-, that any language respects, has not evolved because it illustrates certain laws of nature that govern our physical life and thus our thought and speech. Syntactic rules are necessary for the transmission of meaning and the reason they can infer meaning is because they reflect interactions, organization and dynamism, which is the nature of our world.
While I did not initially find Pinker's arguments for the demands of language having some correlative link to its design to be entirely unpalatable, one particular passage was, to me, particularly circular in what little logic it showed: "There is an obvious advantage in being able to acquire such information about the world second‐ hand: by tapping into the vast reservoir of knowledge accumulated by some other individual, one can avoid having to duplicate the possibly time‐consuming and dangerous trial and error process that won that knowledge. Furthermore, within a group of interdependent, cooperating individuals, the states of other individuals are among the most significant things in the world worth knowing about. Thus communication of knowledge and internal states is useful to creatures who have a lot to say and who are on speaking terms."
ReplyDeleteIf we were to paraphrase, the concluding sentence would become something like: thus communication of certain things is useful to creatures who have those things and can communicate. There is simply a lack of causal reasoning that fails to explain the evolution of language as an advantageous cognitive mechanism. Communication of knowledge and internal states is useful (and would confer an evolutionary advantage): that much most everyone would agree with, I presume. How is it that we have that knowledge and those internal states—with what form, or none at all, of language did they come with? If language did have a hand in developing those things over time, whence came it? Should we take "a lot to say" as meaning a surfeit of internal states, or an unfulfilled need to express? What, exactly, does "on speaking terms" mean in this context, besides arguing for the development of language by its very existence? Putting his rationale in terms of creatures who have "a lot to say" and are "on speaking terms" obfuscates the crux of the argument, which is how precisely those creatures came to be that way (and also begs the question of the chronology of other adaptations cited, such as the vocal apparatuses). When all that is said and done, Pinker's assertion that "it is not necessarily illegitimate to infer both special design and adaptationist origins on the basis of function itself. It all depends on the complexity of the function from an engineering point of view."—supported with examples such as bat sonar that possess anatomical complexity but lack other syntactic or semantic components entirely—is considerably harder to swallow.
We have discussed how language cannot just be a by-product of evolutionary progress. This is an idea I have struggled with consistently. I understand that grammar cannot be just learned and an aspect must be innate due to the poverty of the stimulus argument. What separates us from other animals without a verbal language like ours is our large prefrontal cortexes. This is likely where our capability for language lies. I originally believed that this came to be because language was biologically advantageous. It allowed for a communication that allowed us to off load our cognitive abilities onto other people. This saves time and makes learning safer since we no longer need to test things by trial and error (inductive reasoning). This is to say that we no longer need to eat the poisonous mushroom to figure out it's inedible but rather we can be told and circumvent the danger entirely. This would seem like advantage enough but it was countered that we could actually survive through inductive reasoning and eventually learn to avoid things that are detrimental but we have not just survived, we thrived. There needs to be an additional explanation for why we are able to have interactive speech and this is the aspect I find most troubling.
ReplyDelete"The social value of complex language probably played a profound role in human
ReplyDeleteevolution that is best appreciated by examining the dynamics of cooperative
interactions among individuals. As mentioned, humans, probably early on, fell
into a lifestyle that depended on extended cooperation for food, safety,
nurturance, and reproductive opportunities. This lifestyle presents extraordinary
opportunities for evolutionary gains and losses. [...] The minimum cognitive apparatus needed to sustain this lifestyle is memory for individuals and the ability to enforce social contracts of the form ʺIf you take a benefit then you must pay a costʺ (Cosmides, 1989). This alone puts a demand on the linguistic expression of rather subtle semantic distinctions."
I found this to be a somewhat strange argument to make. It seems to me that there are many other animals have similar lifestyles that depend on cooperation for food, safety, nurturance, and reproductive opportunities. How come humans are considered different from any other animal or insect that engages in cooperation with other beings? I wasn't able to find any explanation for this distinction, which makes it difficult for me to accept the assertion that "this alone puts a demand on the linguistic expression of rather subtle semantic distinctions." One would think, if that were the case, that honeybees, chimpanzees, and ants would display a language on par with human language. Of course, as we learned early on in this class, these animals and insects have ways of communicating that satisfy their needs but they don't have language like humans do.
"Cooperation opens the door to advances in the ability of cheaters to fool people into believing that they have paid a cost or that they have not taken a benefit."
Again, haven't we seen this in other animals that engage in cooperation? If not, how come? It's not clear to me that language would evolve simply to detect cheaters and encourage cooperation. Cheating appears in many instances throughout nature and it's not unique to humans.
"Partial conflicts of reproductive interest between male and female, sibling and sibling, and parent and offspring are inherent to the human condition"
And primates, hyenas, bears... It seems to me that the this section of the paper amounts to "unlike other animals, humans evolved to have language because they depend on the same things as most other animals".
This comment has been removed by the author.
ReplyDelete