Comment by compay | Hacker News Reader

Comment by compay | Hacker News Reader

Parent Full thread

I studied linguistics in the 90's, including a class and a few seminars with Chomsky, but left grad school after a year. At this point I'm neither a linguist nor an expert but maintain a slightly more than passive interest in the field.

These kinds of crackpot articles surface every couple of years in the mainstream press, always from people who demonstrate no knowledge of Chomsky's work beyond having read the back cover of his 1965 "Aspects of the Theory of Syntax."

I didn't read the original article, but as described in the Ars article, the thesis makes absolutely no sense whatsoever. The fact that we may have a hard time reconciling how languages can be diverse in their handling of subject/object/verb order doesn't disprove the concept of Universal Grammar. In fact, it's an argument for Universal Grammar that we can even have a coherent conversation about the way all languages handle subject/object/verb order - Universal Grammar is not about providing a set of rules describe all languages, it's about explaining what a possible rule is, and you have to approach that with logic rather than statistics.

The Ars article talks about "languages that have been evolving for a minimum of 4,000 years," as if those languages simply sprung into existence a little more than 4 millennia ago with no relation to anything else. This is the linguistic equivalent of creationism.

[Edit: expanded acronym "UG" to "Universal Grammar" for clarity]

Sorry, but even though my passing knowledge of linguists may be far outstripped by yours, I still think you may be mistaken.

The statements made by the article are related to something called Universal Grammar, it is not related to syntax theory or Chomsky hierarchies:

http://en.wikipedia.org/wiki/Universal_grammar

According to the article, Chomsky probably didn't invent universal grammar, but he was certainly a strong proponent.

As to the second point, the article claims that languages have been evolving for a minimum of 4,000 years, not a maximum of four thousand years. This is a perfectly accurate statement about all known species of animals. All it really says is that we have records that lead us to know for certain that certain sets of languages had common roots as soon as 4,000 years ago, but we are not sure about their common roots before that.

This sort of argument is very common in genetics; I recall reading a book about how human men can all be traced to a common ancestor 90,000 years back. This does not imply that scientists believe humans sprang into being 90,000 years ago.

Well, I was admitted to a PhD program in linguistics which required more than passing familiarity with the basic concepts of Chomsky's research, of which Universal Grammar is the basic concept.

In my comment UG = Universal Grammar.

I think it's very worthwhile to read the original article in this case as it supports something far less strong than what the Ars article does. It's argument is pretty compelling when you get to what it's actual argument is/should have been. But it hardly topples what it claims to.

Yeah, I tried to be careful to criticize only the Ars article, since I imagine there's a strong possibility the original may have been misquoted or misinterpreted.

Reply on news.ycombinator.com