?

Log in

No account? Create an account
Trevor Stone's Journal
Those who can, do. The rest hyperlink.
Paper Musings 
7th-Dec-2003 11:08 am
Trevor baby stare
The way linguists talk about language acquisition, you get the picture that what's going on is just people inducing a grammar based on token sentences. The Poverty of the Stimulus argument therefore says there isn't enough information available in just the sentences a child hears to learn the grammar.

Suppose we look at language acquisition as learning a grammar and lexicon to express a semantic world. Suppose a child knows (or at least has a good idea) what a sentence means (and this can be determined without language, or built slowly with ungrammatical utterances). Therefore, a significant amount of information about the sentence's meaning, such as the number of objects involved, the sorts of events occurring, etc. is available. Thus, with innateness of basic mentalese and very rudimentary language learning (at the level of lexical binding), grammatical domain specificity may be unnecessary.
Comments 
8th-Dec-2003 04:09 pm (UTC)
Because I love you, I am going to ask this before diving into what I normally do when I see something like this-
Did you want discussion (and all that implies, i.e. "discussion (with an avid linguist who is intensely involved in the Chomskyan theories surrounding mentalese and all the connotations therein)"), or are you just musing for yourself about the paper?
I can go either way. ;)
9th-Dec-2003 12:10 am (UTC)
A little from column A and a little from column B.

I was mostly using the post as a sticky note that can't get lost on my desk, but better educated makes for a better argument.

Note, however, that I'm not trying to say anything substantive about what's actually true about language. The goal of the paper is to analyze the Poverty of the Stimulus argument on philosophical grounds. So in addition to teasing apart what claim nativists make (or ought to), my goal is to show where the argument is open to attack. It can be a good paper if I say "With the right data, empiricists could attack premise 5" even if the right data isn't around.

The insight that inspired my post was of the form "Maybe we aren't asking the right questions." Would it be more helpful to think of natural language as a compiler than a specification?

But if you'd like to dive, you can land in my lap pool any day.
9th-Dec-2003 04:40 pm (UTC)
But if you'd like to dive, you can land in my lap pool any day.

Oooh yeah baby, talk dirty linguistics to me... ;)

I don't know how much help I would be with finding weaknesses of the argument, because even if I did I think my brain would ignore it, and see it as an attack on my main man Chomsky... Besides the fact that most of the time I find his writing to be as watertight as a mermaid's brassiere, so...
This page was loaded Jul 23rd 2018, 3:42 am GMT.