sudanglo:

We do not live in a world for which we can always find right-wrong answers.

As a general proposition, I suppose this must be true, Robb.

But how exactly could the idea that there is a biologically determined grammar engine which all human languages have to conform to be both right and wrong?

Do you imagine that some elements of grammar are universal and the rest we can make up (ie are culturally determined) as we please?

This is how scientific models work: A model isn't right or wrong, but it is apt or applicable within certain bounds. As long as a model -inside its bounds- enables you to make more or less accurate predictions of the future, or it enables you to explain empirical data, then it is apt and applicable. We nowadays know that the mass of a body depends on its speed (relative to a given 0) and hence Newton's laws of motion (v = ∫adt, s = ∫vdt) are theoretically rubbish. This however doesn't invalidate Newtonian physics as a model. When I drive at 60mph for 2 hours, I can still happily use Newton in order to determine how far I have gone. Once I go relativistic with my Hyundai, that's another issue. Same holds for the different models of the atom or subatmoic elements: each model helps us understand the world and make predictions about it. In some contexts some models are better than others; in some context some models can't be applied.

UG is a model for describing and trying to explain some empirical data observed, and for making some predictions of the future. It doesn't matter if this UG "exists", whether it is right or wrong. As long as it helps us with descriptions and predictions, it is a valid model.

sudanglo:Do you imagine that some elements of grammar are universal and the rest we can make up (ie are culturally determined) as we please?

If I recall correctly, what you wrote is exactely what the UG model postulates: UG is a set of unparametrized (and hence acultural) rules. When we acquire L1, we parametrize these rules. We also deactivate some and activate others (all members of the initial set). This model is one way of tackling the problem of the scarcity of input. The unparametrized rule set is culture-agnostic, while the parametrization and activation is culturally determined, in order to use your words.

It would also be naive to think, that once we know the exact "wiring" of our brain, we will know how it works and will be able to predict outcomes. We still have to deal with emergence. In order to predict the working (and the outcome) of a system, it is not enough to know all the rules it is operated by. We cannot predict the behaviour of Langton's ant by looking at its rules. The ant has to dance.

antoniomoya:

sudanglo:Learning a language is like learning the rules of a game.

I have always thought that learning a language is like learning a secret code. A code known only by the speakers of that language.

Amike.

I think it is somewhat like both, and not really like either. And I'm not sure which I'd say it is more like. As a kid, I thought that other languages were just like a code; you switch out one word for another, for example, saying "un perro" instead of "a dog." As I studied Spanish (which I never did learn to fluency), I found that there were things that didn't have a direct counterpart in English, and my perception of languages changed somewhat. But it is still true that in another language, we usually say the same things, that is, express the same concepts, in a different way. But words and grammar aren't the whole picture. In a way, learning a language is like learning the rules of a game, because we have to learn a different system of expression. On Lang-8 I often deal with entries in English that are not grammatically wrong, but are just not "the way it's said." In any language, even Esperanto, there is this aspect of what is normal to say, and what is grammatically correct, but sounds stiff or foreign. In any such analysis we risk oversimplifying the issue.

It is not my understanding that people like Pinker are saying that UG is a model. Rather it is something that we inherit as a species - something that actually exists, like the neuronal circuits in the visual system of some animal brains that respond to edges or vertical lines.

As soon as you say that all the rules pre-exist and some are activated and some not depending on the language acquired, then the 'model' seems to me to be lacking in explanatory force. That's almost like saying UG exists by definition. Whatever the rule it conforms with UG.

How does this help us make descriptions or predictions?

sudanglo:It is not my understanding that people like Pinker are saying that UG is a model. Rather it is something that we inherit as a species - something that actually exists, like the neuronal circuits in the visual system of some animal brains that respond to edges or vertical lines.

As soon as you say that all the rules pre-exist and some are activated and some not depending on the language acquired, then the 'model' seems to me to be lacking in explanatory force. That's almost like saying UG exists by definition. Whatever the rule it conforms with UG.

How does this help us make descriptions or predictions?

I am not sure. I was never much into language acquisition. I think the central motivations for UG were the scarcity of input, the fact that L2 acquisition is completely different to L1 acquisition, and the structural similarities between languages. Basically that during L1 acquisition something irrevocable happens in (or to) our language system. That's why even if we speak a foreign language for 90 years every day, it still is a foreign language. No idea though.

Maybe some scholars have taken UG too literally, as e.g. in optimality theory. But if one denies an instinct for language, then how does one explain the fact that all babies babble and that newly-borns react more positively to a language spoken in their environment at a prenatal stage than a language that was not spoken around them? And how does one explain that all humans (devoid a brain deficiency) do indeed acquire L1? There are the cases of two children raised together but separated from other language speakers in China and the hearing impaired children of Nicaragua who invented their own sign language.
Of course, we can question the meaning of instinct. Then with which word does one describe the reaction of a duckling raised without a mother to water? To the reaction of a lone kitten to mice and fledglings? Etc. etc.? Maybe then a different word is needed to describe the human tendency/drive to speak. But that it is there cannot be denied.
Of course, we cannot discuss a language without nouns and verbs because the concept does not exist in our language which binds us. But then no language yet discovered among humans lacks nouns and verbs. Adjectives many be verbal and adverbs may only exist as phrases, but the basic concepts are there. Is this all just happenstance?

robbkvasnak:because the concept does not exist in our language which binds us

+1