Tuesday, April 3, 2012

Everything was a problem and we did not understand a thing

"Everything was a problem and we did not understand a thing": Beautiful quote, too bad Noam Chomsky is using the academic "we" and actually means "you." It would be nice if he could manage to concede that his universal grammar theory, the drum he's been banging for decades, might have some flaws or failings; surely he could be forgiven that, linguistics was a young science when he got into it. But nope: Any evidence that contradicts his theory "can't be true," not possibly:
Let's start with the idea that everyone connects you with from the 1950s and ’60s—a "universal grammar" underlying all languages. How is that idea holding up in 2012?  
It's virtually a truism. There are people who misunderstand the term but I can't deal with that. It's perfectly obvious that there is some genetic factor that distinguishes humans from other animals and that it is language-specific. The theory of that genetic component, whatever it turns out to be, is what is called universal grammar.  
But there are critics such as Daniel Everett, who says the language of the Amazonian people he worked with seems to challenge important aspects of universal grammar.  
It can't be true. These people are genetically identical to all other humans with regard to language. They can learn Portuguese perfectly easily, just as Portuguese children do. So they have the same universal grammar the rest of us have. 
"Whatever it turns out to be" = basically, "you figure it out numskulls."

There's something about his idea that "there is some genetic factor that distinguishes humans from other animals"  that feels so racist. Right? It's like, Blech, filthy monkeys! Can scarcely utter a complete sentence...

And it's not at all obvious to me that this "genetic factor" is "language-specific." Couldn't we just as easily say:

  • "It's perfectly obvious that there is some genetic factor that distinguishes humans from other animals and that it is plumbing-specific." 
  • "It's perfectly obvious that there is some genetic factor that distinguishes humans from other animals and that it is cheese-specific." 
  • "It's perfectly obvious that there is some genetic factor that distinguishes humans from other animals and that it is Nintendo-specific." 
Bees and birds communicate in ways that is language-like. Maybe their toasters just don't look like ours.


  1. The thing about that interview is that he comes off as glib, and perhaps he was, but the nature of his argument (that is, the reason behind the theory that is Universal Grammar) is relatively sound.

    A better way of phrasing it might have been:

    There are certain aspects to first language acquisition--like the mistakes that children make versus the mistakes that children do NOT make (e.g., morphemical mistakes are often made, such as gooses, bringed, typer (instead of typist); whereas syntaxical errors are not)--but all of that is tied up in the issues of neurolinguistics. So you wind up with a person who is cut off from speech and their brain's plasticity reaches the point that they can no longer acquire an initial language with the anything remotely resembling ease.

    But that's an extreme exception.

    The situation regarding Daniel Everett's research is dicey because he's one of "only a dozen" people who speak Piraha, but the crux of his argument is that Piraha doesn't have recursion.

    (NOTE: Recursion is the nesting of clauses and one of the things that makes language so fancy is that a person could write a sentence into infinity and it would still be, technically, grammatically correct.)

    One of the tenets accompanying Universal Grammar is that all languages are equally complex, so if there is a language that doesn't have recursion, then it could be argued that it's an example of a language that lacks a specific, complex structure that other languages have. That doesn't write it off as a simple language or basic language, it's just a case of a circle among squares, as it were. The two problems in the cyclical debate that's surrounding Everett's research, however, is that Everett is not only the expert in Piraha, but also it's not exactly clear whether or not recursion is required as part of UG.

    Chomsky may or may not have written a paper in 2002 that touches on recursion, but the supporters of UG are saying that the target audience for this paper was the general reader who isn't necessarily schooled in linguistics (see the argument, early middle of the page: http://chronicle.com/article/Researchers-Findings-in-the/131260). In other words, much like the interview at Slate, the paper simplified matters which need to be formalized in order to make linguistic sense.

    With all of that said, I do get your argument, and I get Chomsky's, and actually, I even get Everett's. It's entirely possible that Piraha has a structure which functions differently than the other languages which have been studied up until now. If that's the case, the real problem is how we go about studying it alongside the system we currently have in place. It's very difficult to start from scratch. But great changes in any field always prove difficult. It's just a matter of being inventive enough to think through them.

    1. My problem with Chomsky's theory is that it ends up feeling like it's not a theory, i.e., something that could be disproven -- it's so abstracted that it seems like he could always argue "that's not what I meant."

      Functional grammar, as a general way to approach linguistics, just makes far more intuitive sense to me.

      And even generally, I think we tend to shove explanations off on genetics to an unnecessary degree.

      I find his glibness w/r/t politics amusing of course, because I don't think he so far off base. :)

  2. We do depend on genetics a lot, yes. But what they're most afraid of is ripping down our current foundation and having nothing to stand on in its place.

    1. If you mean linguistics specifically, I think it's sort of a Chomsky-perpetuated myth that there is no other option or foundation. The majority of linguistics programs in the US teach Chomskyan generative grammar as though it were the accepted standard, but in actuality it's not.

  3. American Sign Language (to take one example) is a grammatical, syntactical language. Studies have been done of babies who were born deaf, and both of whose parents were deaf, and both of whose parents communicated with American Sign Language; the studies have found that the babies acquire American Sign Language in the same manner, and at roughly the same pace (on average) as babies who can hear typically acquire oral/auditory languages.

    For instance, deaf babies of deaf parents, at the age of a few months, will commonly start making simple signs or portions of signs, and commonly the same several signs from one baby to another -- more or less the equivalent of oral-speaking babies commonly starting out saying "ba" and "ma" and "da" and so on. More complex signs (like more complex sounds and syllables) tend to come later.

    There are documented instances of scientists teaching chimpanzees and gorillas to use sign language for limited simple communication. And in some instances, scientists have observed chimps and gorillas stringing signs together to make something resembling rudimentary sentences.

    Not to exaggerate this. But, to my thinking, examples such as the above raise some intriguing questions about whether, and how much, language may or may not clearly distinguish humans from other animals.

    This aside from questions regarding what role genetics may or may not play in the use and acquisition of language.

    1. It seems like we are always learning about things that animals can do that we assumed they couldn't.

      Chomsky's theory seems predicated on an assumption that there are things humans can do (namely language) that animals just can't, but to me (and others) it seems more a matter of degree.

  4. I hate to quote from Wikipedia, but don't want to search around for another high-level summary. Here's some background on differing viewpoints on language evolution/acquisition>>>>>

    Noam Chomsky is a prominent proponent of discontinuity theory, and on this issue he stands quite isolated among his academic peers. He argues that a single chance mutation occurred in one individual on the order of 100,000 years ago, triggering the 'instantaneous' emergence of the language faculty (a component of the mind-brain) in 'perfect' or 'near-perfect' form. The philosophical argument runs, briefly, as follows. Firstly, from what is known about evolution: any biological change in a species arises by a random genetic change in a single individual, which then spreads throughout its breeding group. Secondly, from a computational perspective on the theory of language: the only change that was needed was the cognitive ability to construct and process recursive data structures in the mind (the so-called property of "discrete infinity", which appears to be unique to the human mind). This genetic change, which endowed the human mind with the property of discrete infinity, Chomsky argues, essentially amounts to a jump from being able to count up to N, where N is a fixed number, to being able to count indefinitely (i.e. if N can be constructed then so can N+1). It follows from these assertions that the evolution of the human language faculty is saltational since, as a matter of logical fact, there is no way to gradually transition from a mind capable only of counting up to a fixed number, to a mind capable of counting indefinitely. The picture then, by loose analogy, is that the formation of the language faculty in humans is akin to the formation of a crystal; discrete infinity was the seed crystal in a super-saturated primate brain, on the verge of blossoming into the human mind, by physical law, once a single small, but crucial, key stone was added by evolution. [6][7]

    Continuity based theories are currently held by a majority of scholars, but they vary in how they envision this development. Among those who see language as being mostly innate, some — notably Steven Pinker[8] — avoid speculating about specific precursors in nonhuman primates, stressing simply that the language faculty must have evolved in the usual gradualistic way.[9] Others in this intellectual camp — notably Ib Ulbaek[10] — hold that language evolved not from primate communication but from primate cognition, which is significantly more complex. Those who see language as a socially learned tool of communication, such as Michael Tomasello, see it developing from the cognitively controlled aspects of primate communication, these being mostly gestural as opposed to vocal.[11][12] Where vocal precursors are concerned, many continuity theorists envisage language evolving from early human capacities for song.

    1. Also, Chomsky apparently didn't have much respect for kittens>>>

      It is a popular misconception that Chomsky proved that language is entirely innate and discovered a "universal grammar" (UG). In fact, Chomsky simply observed that while a human baby and a kitten are both capable of inductive reasoning, if they are exposed to exactly the same linguistic data, the human child will always acquire the ability to understand and produce language, while the kitten will never acquire either ability. Chomsky labeled whatever the relevant capacity the human has which the cat lacks the "language acquisition device" (LAD) and suggested that one of the tasks for linguistics should be to figure out what the LAD is and what constraints it puts on the range of possible human languages. The universal features that would result from these constraints are often termed "universal grammar" or UG.

    2. I think it works especially well if you imagine Noam talking in the Werner Herzog voice: "As you can see this kitten appears to lack even the most basic grasp of language"

    3. "As you can see this kitten appears to lack even the most basic grasp of language"

      And therefore, on the hierarchical linguistics food chain (also known as the Linguistics Food Pyramid, not to be confused with the Linguistic Hierarchy of Needs) we feed on kittens.

    4. Although our daily allotment of kittens is not to exceed 3.

  5. I am always skeptical of arguments that invoke human exceptionalism. Good post, Elisa.

  6. Actually, I meant systems overall. It's a problem across all fields that require a form of watering-down of concepts for public consumption. It's just as easily applied to government. The reason we can't currently consider anything outside of bipartisanship is because the average voter doesn't know enough about our current structure (consequential of the watered down information we're given due to a complicated system) to be able to fully grasp splitting from it.

    So, going back to linguistics, because Chomsky is a household name in the field (and even that's a stretch), the information most of the general public receives is based on his work and that's as far as they care to know, since most people don't bother digging beyond the information they're given. To that end, the idea of splitting from something that is already so complicated seems unfathomable. It's a case of Refusing to Fix That Which Is Seemingly Unbroken.

    1. Yes, definitely. Consensus is extremely powerful, and even when a consensus opinion loses grip among scientists, it often keeps its hold on the general public for years/decades. People don't like revising their opinions, it's a pain.

  7. I'm so glad I quite grad school before I learned out to dismiss ideas by reciting a magic spell such as "It can't be true" when confronted by a differing opinion or theory.

    Then again, I've made an enjoyable hobby out of revising my opinions when presented with new evidence.

    As for a universal grammar that separates us from the rest of the animals, I'll argue that it might separate us from some animals. Not all. Clearly, gorillas can be taught to communicate with humans via sign language.

    As for kittens, Koko the gorilla and Chomsky could discuss them at length.

    Koko: I like my kitten.
    Chomsky: She can't speak. I have a Ph.D, so I notice these things.
    Koko: The hipsters have moved on and don't care about you any longer. Pet this kitten.

    1. The ability to change your mind in public is a sign of genius!

  8. I just read a book about this, so I'm practically a world-recognized expert on the topic (for the moment, until it recedes back into the haze). First, there must have been some complex ability that was "exapted," or adapted to communication. The best candidate is the throwing faculty, which synchronizes the motion of shoulder in relation to upper arm in relation to forearm to hand to fingers etc. etc. The idea is that each sentence is constructed on the fly, by adding new elements, just as a thrower calibrates the motion of each segment of the throwing arm on the fly. I realize that's weak, but something like that. Language then spread so quickly by hopping on the back of reciprocal altruism, which afforded huge competitive advantages. I wouldn't recommend the book, but there is a book by Bickerton, Adam's Tongue, that is a blast. He has a very impish sense of humor and does a great job of ridiculing the competing theories, such as the idea that language communication (basically "climb! leopard coming!") could ever evolve into human communication. FWIW, I'm definitely on the "Chomsky must be right, there's no point in arguing this" side, which is probably what irritates you.

    1. Why must he be right, though? No one has ever once said anything remotely convincing to me in that regard.

      Hilarious comment on the original article: "As I understood it from my study (which is limited to reading Guns Germs and Steel)..."

  9. There's also this to consider: in human beings, the larynx is positioned further forward (or higher up) in the throat than it is in most animals, including other primates. I'm not sure why this occurred -- may have been part of a general reshaping of parts of the anatomy as apes began to stand up erect.

    And the higher/further-forward placement of the larynx is a crucial element in enabling the highly varied and finely articulated speech of human beings. Merely making complex sounds isn't, of course, the same thing as making language -- many animals make highly complex sounds (sometimes too fast or too high or low for us to hear the full range of the sounds). And language isn't always or necessary vocal.

    But it's interesting to speculate about what may have become (or not become) of human language if humans were only capable of making a more limited number of sounds, similar to other primates, whatever our cognitive abilities might have been.