Slog

Slog Music

Music, Nightlife,
and Drinks

Monday, March 26, 2012

What Language Is

Posted by on Mon, Mar 26, 2012 at 8:35 AM

This...

Language is possible due to a number of cognitive and physical characteristics that are unique to humans but none of which that are unique to language. Coming together they make language possible. But the fundamental building block of language is community. Humans are a social species more than any other, and in order to build a community, which for some reason humans have to do in order to live, we have to solve the communication problem. Language is the tool that was invented to solve that problem.

...one of the best definitions of human communication I have come across was made by a man who entered linguistics by way of a book that many believe was authored by a supernatural and super-powerful ape, the Bible. Daniel Everett, the man in question, began with something that deserves derision (learn the language of a Brazilian tribe for the purpose of translating a book about an all-knowing ape) and ended up with something that deserves consideration: A strong argument against Noam Chomsky's universal grammar—Indeed, "with indirection find direction." Everett, however, is a figure under heavy fire for reasons that do not seem unreasonable.

 

Comments (12) RSS

Oldest First Unregistered On Registered On Add a comment
1

Do you now define language as "urban"?

Why not...everything else is.
Posted by Supreme Ruler Of The Universe http://_ on March 26, 2012 at 8:40 AM · Report this
2
The mere interaction in creating a translation of any language will, by it's very nature, create a tainted result. What ever initial language and cultural programing that exists in the translator can not be stripped away. This taint causes the intermediary, otherwise known as the translator, to fail at being a true two way bridge. Multiple translators can weed out the discrepancies, but even with numbers they still will never be perfect.

The Bible is actually an excellent example of this. Thousands upon thousand of translations back to the original scrolls and yet no version of the english Bible is identical to another.
Posted by kmq1 on March 26, 2012 at 8:55 AM · Report this
3
Everett's ostensible problem for Chomsky stems from a pretty simple misunderstanding of the claims made about language. Everett offers evidence that this language does not display certain constructions and these constructions are taken to be indicative of some fundamental properties of the human language system. Given this, he says that the speakers of this language must lack the hypothesized fundamental properties. If a language lacks thes 'fundamental' properties, then, contra Chomsky, they really aren't fundamental to begin with.

Assuming the evidence to be correct, this argument does not go through. Chomskyan claims about language are basically of the following type: There is a shared capacity for language among humans and that this capacity has certain properties. Chomskyans go on to try to study what those properties are. Nowhere in their claim is it suggested that every language displays every property. Such a thing would obviously false: English doesn't have the case system that Finnish does. That doesn't mean that the same underlying computation is not responsible for both languages. Everett seems to think this is what is meant though.

Rather, Chomsky makes a more interesting claim. Languages can lack whatever they want there is no lower bound to what a language can do (higher power entails lower power). What Chomsky claims is that there is an upper bound to what a language can do. There are concepts or relations that our brains are perfectly able to do generally, but language is constrained in an interesting way. One example, In English when forming a yes-no question, we front a verb-like element:

I should leave --> Should I leave

It's not necessarily the first verb-like element:

The man who must eat should leave --> Should the man who must eat leave?

but NOT: Must the man who eat should leave?

That is, it seems that language deals in hierarchical structure, not linear order. As such, Chomsky would predict that no language displays constructions that depend on counting or other linear depedencies. The fact that Everett found a language that lacks something is pretty boring. In English we can say things like Ivy's sister's friend, but the same is not possible in German. This does say anything about Germans' capacity for language.

If Everett were able to find a language that say, made yes-no questions by always moving the 3rd verb to the front, then Chomsky would be in trouble. This is logically possible, but very unlikely

More...
Posted by bradl on March 26, 2012 at 10:15 AM · Report this
seandr 4
@3: If Everett's data are so boring and inconsequential, why are Chomsky's drones trying to take him down with specious claims of racism and fraud?

Posted by seandr on March 26, 2012 at 10:22 AM · Report this
5
Well, I can't speak to the particular motivations for those people, but I can make a guess as someone in the field.

I don't think the misunderstandings are one-sided. The people making those claims are missing the point too. I think you're right to call them drones, not because they unthinkingly follow whatever Chomsky says but rather because they just don't do much critical thinking in general. Every field has such people and when they are loud, it makes the rest of us look bad. The bristle at the Everett's conclusion and default to the ad hominem. It might also be the case that some people understand the flaw in the reasoning yet get angry enough to act foolish. Everett's misunderstanding is a big one and not even an uncommon one. It gets me angry sometimes as well because of the hoopla surrounding it.

Also, I don't mean to say that his data are boring or even inconsequential. This is a fascinating language and well worth studying. But as in most sciences, what is consequential and exciting is so for only a niche group. I think the Piraha stuff is of this sort.
Posted by bradl on March 26, 2012 at 10:37 AM · Report this
6
@4 - I think while their reaction belies their adherence to their preferred theory, rather then the truth, whatever it may be, i do think they have some valid points, such as, how many people besides Everett even speak Piraha? Very few, making a lot of his claims difficult to test. Its like they've all been peer reviewed in his own mind. I think its possible that Everett is wrong, AND that the Chomsky camp is not open minded on the matter.
Posted by longball on March 26, 2012 at 11:14 AM · Report this
7
@2

This. A thousand times this.

The Iliad is, for example, a story with which I am very familiar. I've read Fagles' translation, Lattimore's translation, and its entirety in Greek. And I have to tell you, the problem of a translation is that it essentially produces a simulacrum; you're not reading Homer's Iliad when you read the Fagles' translation; rather you're reading Fagles' Iliad.

Dactylic Hexameter doesn't translate well to English; so the poetic metric is not only frequently, but completely lost in translation. The power and sense inherent in the Greek lexicon also can simply not be translated perfectly into English; one will always produce different, false impressions not intended in the original.

To truly get a sense of the original, you have to read it in its own tongue; but even that has limitations; having learned Greek through the lens of my knowledge of comparatively similar grammatical and lexical systems has certainly colored my knowledge of Greek say differently than a native speaker (if anyone can be said to be so nowadays) or a native speaker of a different language--like German.
Posted by Central Scrutinizer on March 26, 2012 at 11:52 AM · Report this
gwhayduke 8
There are more. It is difficult to accept the veracity of this excerpt and, by extension, article. Language is not unique to humans. A mound of evidence almost certainly contradicts this claim.
The cognitive skills necessary -- and, indeed, employed for language -- have been observed and subjected to detailed study across the animal kingdom. The most famous non-humans are most (all?) primates, elephants, and dolphins. There are more.
Posted by gwhayduke http://www.farmsanctuary.org/videos/celebrity-ambassadors/ellen-degeneres-shares-why-she-supports-farm-sanctuary/ on March 26, 2012 at 1:48 PM · Report this
9
@8

Saying that language is or is not unique to humans depends on how you define language. If all one means by language is a systematic mode of communication then it is correct to say that even unicellular organisms employ language.

This is not to say that all of these systems have the same expressive power or are qualitatively the same. Human language, as opposed to all other animal communication systems have, among others the following traits:

Arbitrariness: Aside from onomonopia, a word or sentence's form is independent of its meaning. This is vanishingly rare in animal communication

Discreteness: There are no 10.3333 word complete sentences. Again, not to be found in animal language. Instead animal communication is carried out in an analogue, continuous fashion.

Displacement: It is possible to make propositions about the future or the past, both impossible ones and impossibles: "If I had been Lincoln I would have stayed home and watched TV instead"

There are others. This is not to say that animal communication is less worthy than that of humans or that the human capacity for language is better. It is simply radically different. Animals are massively impressive (Read about bee navigation and you'll find that they are doing really complex math when foraging), but there are things that are proprietary to humans. These proprietary things are what linguists study.

Posted by bradl on March 26, 2012 at 3:14 PM · Report this
10
"If Everett's data are so boring and inconsequential, why are Chomsky's drones trying to take him down with specious claims of racism and fraud?"

Answer: they're not. If you re-read the article, you'll see that the only evidence is Everett's accusation. There's no text from "Chomsky's drones" calling him a racist, and as for being a fraud - what exists is 50 pages of densely argued evidence in a top journal that his claims are not supported by the data (and he is misinterpreting the opposing views too).
Posted by Linguist Drone on April 5, 2012 at 3:46 PM · Report this
11
"If Everett's data are so boring and inconsequential, why are Chomsky's drones trying to take him down with specious claims of racism and fraud?"

Answer: they're not. If you re-read the article, you'll see that the only evidence is Everett's accusation. There's no text from "Chomsky's drones" calling him a racist, and as for being a fraud - what exists is 50 pages of densely argued evidence in a top journal that his claims are not supported by the data (and he is misinterpreting the opposing views too).
Posted by Chomsky's Drone on April 5, 2012 at 3:50 PM · Report this
12
bradl makes the following argument about the nature of language:

One example, In English when forming a yes-no question, we front a verb-like element:
I should leave --> Should I leave
It's not necessarily the first verb-like element:
The man who must eat should leave --> Should the man who must eat leave?
but NOT: Must the man who eat should leave?
That is, it seems that language deals in hierarchical structure, not linear order.

-------
The problem with this argument is that it assumes that speakers (and presumably listeners?) begin with a declarative sentence and then form a yes-no question by "moving" and "transforming" its constituents. "Should" is moved backwards in time against the speech stream. Chomskyan theorists typically say that this is an "abstraction." However, these abstractions make about as much sense as the emperor's invisible new clothes. Acquisition researchers like Tomasello show that questions and statements are both acquired between 2 and 3 years of age, before children use any substantial syntax. And syntactians like Ginzburg and Sag (see their 2000 book Interrogative Investigations) show that extremely precise theories of syntax do not need to use transformations to explain these phenomena. Once you abandon the transformational metaphors that have clouded Chomsky's thinking, Daniel Everett's book makes sense and you will wonder why Chomsky has been so misguided.
Posted by James4321 on May 1, 2012 at 1:04 PM · Report this

Add a comment

Advertisement
 

Want great deals and a chance to win tickets to the best shows in Seattle? Join The Stranger Presents email list!


All contents © Index Newspapers, LLC
1535 11th Ave (Third Floor), Seattle, WA 98122
Contact | Privacy Policy | Terms of Use | Takedown Policy