An Interesting Article on Language and Thought

Here is an interesting article on language and thought. What intrigued me was the following:

“Dr. Chomsky proposed that there is a universal grammar for all human languages.... The search for linguistic universals yielded interesting data on languages, but after decades of work, not a single proposed universal has withstood scrutiny.”

For some time now, I have thought that Chomsky could not possibly be right about the existence of a universal grammar. Such a grammar would help infants learn a language (via innate knowledge), but languages just seem too varied for there to be any rules applying to all.

The example I’ve been thinking about the most is number. In Japanese, according to a friend, there is no distinction in form between singular and plural. In English, we generally distinguish between singular and plural, and we typically do so by adding an s to the end of the word, though there are plenty of exceptions. Spanish and Portuguese are similar. German is more complicated, but the most complicated setup I’ve seen is in Arabic. Here are the highlights:

1. There are many different plural patterns. (One of my grammar books lists fifteen and makes no pretense that these are all of them.)

2. When you refer to two items, you must use neither the singular nor the plural, but something called the dual. Oddly enough, everyone is familiar with two examples of this, for “Taliban,” strictly speaking, means “two students” and “Bahrain” means “two seas.”

3. The plural is used for at least three items but no more than ten. With eleven or more items, the singular is used.

4. Some nouns have a collective form, in addition to all the others. For example, the word for one fish is samaka, two fish is samakan, three to ten fish is asmaak, but fish collectively is samak. Notice that the collective form is actually shorter than the singular form.

With all the different ways that number is dealt with in various languages, it seems to me that the “rule” for number is nothing other than “whatever you find,” which of course is no help at all to any infant learning a language.

Comments

[wheels]

Chomsky's ideas - deep structure and the like - have always sounded incorrect to my intuitions. Not very rigorous, I know, but from my knowledge of languages, I've always felt that Sapir-Whorf seemed likelier to be correct.

James Cooke-Brown, who started developing Loglan as an effort to test the strong form of the Sapir-Whorf hypothesis, has used as examples of language differences that seem to support Sapir-Whorf the tenseless nature of the Hopi language, and the language of the Tobriand Islands that treats individual things as specific instances of the collective. I don't remember his specific examples for that one, but I recall that he anthropomorphized them by saying, for example, that a glass of water is treated as an instance of "Mr. Water," and a rabbit is an instance of "Mr. Rabbit," referring, respectively, to "all the water in the world," and "all the rabbits in the world."

I think any deep structure that can account for all variants of human language has to contain very little actual structure, personally.

[John Pepple]

Thanks for the input, especially the examples. And be sure to follow the link for some more unusual examples. I certainly agree with your last sentence.

[Account Deleted]

Deacon's The Symbolic Species has a lot more to say on Chomsky and the concepts discussed here. For fundamental info, try also Langer's Philosophy in a New Key. In a nutshell, we speak because our brains require it, and very young children have made our languages what they are -- so the languages could be learned by humans.

[John Pepple]

I've never heard of Deacon's The Symbolic Species. I'll have to put in on my list of books to read. Thanks.