Talk:Noam Chomsky

From RationalWiki
Jump to: navigation, search

Contents

[edit] Liberal?!?!?!

Chomsky is hardly by any stretch of the imagination a liberal, unless you're a Conservapedian who defines "liberal" as "not conservative." PFoster 20:08, 1 November 2007 (EDT)


I've removed the bit about Chomsky being "popular and well known in almost every country of the world except his home country of the United States," and that "he reason for this is that the powers to be in America don't like what Noam has to say so they use their propaganda arm, the mainstream media, to marginalize Professor Chomsky and shut out his opinions" because 1. a lot of the language is vague, 2. the statements are completely without any documentation to support them and 3. I'm not sure how "marginalized" Chomsky really is - he's Professor Emeritus at one of the most reputable universities in the country, and I can buy each and every one of his books at the Barnes and Nobles or the Borders at the mall. PFoster 15:38, 15 November 2007 (EST)

[edit] Hugo Chavez

The comment about Hugo Chavez and Chomsky's book seem to be true-ish. http://www.nytimes.com/2006/09/23/books/23chomsky.html

I don't know if one can conclude that Chomsky has become increasingly popular due to this though. I'll leave it out as it's not really a very important point about Chomsky.JoeDuffy 08:57, 16 April 2009 (EDT)

I changed the article to reflect the story above. CreamTwoSugars 09:33, 16 April 2009 (EDT)

[edit] His linguistics is not so good.

Computer programs based on Chomskian linguistics are poor translators, statistics-based devices translate much better. Modern neurology shows that real brains are based on statistics, which they must to be able to circumvent unplanned obstacles like those commonly found in nature (which robots steered by classical computers cannot). It has also been shown that statistical processing makes it possible to infer all grammatical rules from infant-directed speech without innate preconceptions, and that all universal or near-universal grammatical rules can be explained as innovations out of linguistic laziness without any need to invoke any common instinct(Francisco Lacerda) and that it is possible to develop/recover any cognitive ability, including language, after brain damage destroying any specific part of the brain, provided just that there is no rigid curriculum to prevent unconventional approaches to learning (Mind, Brain and Education).95.209.107.197 (talk) 14:08, 13 November 2011 (UTC)Martin J Sallberg

I don't know of any computer programs that attempt to use chomsky linguistics to form the auto translation - and in fact Chomsky himself would say a computer cannot (yet) do language, so it would not be a basis for translations. Brains are not based at all on statistics, they are based on a process of interlaced awareness - that is, linked memories, linked assumptions, and linked creativity which produce novel solutions to novel problems. "statistical processes make it possible to infer grammatical rules?" are you kidding me? if that were teh case, we would have computers that could understand language. we do not, and so far we cannot. Lacerda does not, as far as I can tell, talk about "linguistic laziness", but rather stresses the way the mind reinforces language into simple patterns of sound. I assume you've studied ASL, yes? Cause that should strongly challenge your "statistical rule" theory. Further, I'm not sure that I've seen studies that actually support the idea that parts of the brain do not control language, or that they have been rewired, for all it is an intriguing possibility.--Pink mowse.pngGodotI smell roasted chestnuts. droollllllll. 15:29, 13 November 2011 (UTC)
When I wrote that brains are based on statistics, I did NOT deny the importance of connectivity, I just meaned that neurons process information in a non-binary way that allows gradual comparisons instead of absolute yes and no. And yes, statistical computers are fairly good translators. Statistics can reveal any correlation patterns and do not need to test every rule separately. It is just that inferring the rules in the grammar do not mean understanding what the language means, because no "disembodied brain" (intelligence without sensory input) can have any frame of reference to learn what the words mean. Another contributing limitation is that in non-binary brains, brainpower is all about improving neurological precision. It is often said that human language differs from animal communication by having specific words instead of just signals for whole contexts, which is mostly correct with some modifications for brainy animals like apes having intermediate levels of neurological precision (more specific than normal animal signals but less specific than proper human words), meaning that a statistical computer with much less capacity than a human brain should, by statistical language theory, have animalistic language limitations. Maybe I should call it "interferometric brain" instead of "statistical brain". Consider that during sleep, different parts of the brain do not communicate with each other. Voices heard in dreams speak complete nonsense, just like non-recovered aphasia patients. That some parts of the brain show more activity than others during specific tasks do NOT mean that part of the brain does the task alone, only that that task activates a statistically higher number of connections passing through that brain region. Destroying such regions may cause seemingly specific impairments, but it is just because of the number of connections involved, NOT because some regions is anyhow specialized for tasks others cannot learn (such iron-clad specialization is impossible because non-binary brains act by pattern emergence from examples and therefore cannot treat instinct-information fundamentally different from how they treat sensory information) And yes, there several documented cases of people recovering after having parts of the brain destroyed. Brooke had almost his whole left hemisphere removed at age 11. He could not speak just after waking up from the narcosis, but after six weeks he could say "mama", after one year he could have a decent conversation, and barely three years after the surgery he could attend a normal school class. It is all described in the show "Mind, Brain and Education". And yes, the metastudy shows that the key factor in recovering is a tolerant, flexible environment. The show contrasted the plasticity of the brain itself with the implasticity of the peripheral nervous system, remarking that the origin of dyslexia is in the sensory cells such as the retina, which I am sure is because the brain is a lump, allowing easy formation of new connections, while the peripheral nervous system consists of linear threads incapable of horizontal communication.

79.138.208.52 (talk) 17:44, 20 November 2011 (UTC)Martin J Sallberg

I don't know anything about computer translators, but you're wrong that neurons process information in a non-binary way. They operate on the all-or-none principle. When you say that the brain is "statistical" and that information is processed in a non-binary way, do you mean a Bayesian-style connectionist model? I think you're confusing some terms here. Nebuchadnezzar (talk) 18:14, 20 November 2011 (UTC)

Just because neurons either fire or not fire do ABSOLUTELY NOT mean that they always fire at the same intensity when they fire or that other neurons react as if it were the same intensity. In fact, neurons can and do fire at different intensities, and that variation affects not only whether or not the receiving neurons fire in turn, but also at which intensity they fire if they do fire. Furthermore, there are synapses of different degrees of permeability, some let almost the whole signal through but there is a scale with more and more weakening of the signal for some synapses. The evolutionary reason why brains must be non-binary and statistical is that nature is full of unplanned obstacles, which robots with binary steering units cannot circumvent. 109.58.98.10 (talk) 15:55, 23 November 2011 (UTC)Martin J Sallberg

[edit] Regular Expression

Regular Expressions were invented by Stephen Cole Kleene before Chomsky worked on the topic. Chomsky only invented the Context-Free and Context-Sensitive language classes.--141.70.13.66 (talk) 18:53, 3 January 2012 (UTC)

[edit] Conspiracy theories

I don't know much about Prof. Noam Chomsky, but I have seen his work used a lot by conspiracy theorists. I know for a fact that he has denounced 9/11 truthers, but does he ever mention in his words anything of the likes of "New World Order" or such? Mr. Anon (talk) 18:50, 15 April 2012 (UTC)

Not really in the way you're looking for. He doesn't go about arguing for secret cabals or hidden political formations; he's looking at political power and class interests. Some people read that as a conspiracy theory in itself ("the NTY is in the back pocket of the Republican Party/big business/etc. etc.") but basically he sees power as a structural phenomenon, in which business interests, politics, the media, etc all play their parts. P-Foster Talk "The existing superstructure has handed out crumbs. We don’t want crumbs; we want the whole loaf now.” --Ras Frank I 19:08, 15 April 2012 (UTC)
Personal tools
Namespaces

Variants
Actions
Navigation
Community
Toolbox
support