Anonymous at Wed, 22 May 2024 20:34:20 UTC No. 16188477

>>16188268
Linguistics is very interesting field. If I was to study any humanities, it would be linguistics

🗑️ Anonymous at Thu, 23 May 2024 04:51:39 UTC No. 16188978

>>16188268
No only it got thoroughly debunked by machine learning, it even offers a rather straightforward way how language can be learned. The neocortex works as an autoencoder that filters the world into a small number of labels, and so it does with speech. Then some kind if (cross) correlation happens that connects the labels extracted from the world with the labels that are extracted from speech.

Anonymous at Thu, 23 May 2024 05:10:53 UTC No. 16188985

>>16188268
No only it got thoroughly debunked by machine learning, it even offers a rather straightforward way how language can be learned. The neocortex works as an autoencoder that filters the world into a small number of meaningful features, and so it does with speech. Then some kind of (cross) correlation happens that connects the labels extracted from the world with the labels that are extracted from speech.

Anonymous at Thu, 23 May 2024 06:07:27 UTC No. 16189026

>>16188985
>No only it got thoroughly debunked by machine learning

How? There's little evidence that humans learn language the way machines do, even if machines can learn language to a reasonable degree now.

Anonymous at Thu, 23 May 2024 06:10:43 UTC No. 16189033

>>16189026
Machines don't learn, they just refer to a lookup table as per their programming

Anonymous at Thu, 23 May 2024 06:13:42 UTC No. 16189035

>>16189026
I'm describing the way people could (and likely do) learn it.

Anonymous at Thu, 23 May 2024 06:14:46 UTC No. 16189037

>>16189035
How does that 'debunk' linguistics though? Like, what conjectures or ideas are debunked by this?

Anonymous at Thu, 23 May 2024 06:22:27 UTC No. 16189046

>>16189037
It's debunked by the fact that it's not even alike.

Anonymous at Thu, 23 May 2024 06:27:39 UTC No. 16189051

>>16189037
Like this, here are several features of a latent space (people are much better at this). Then you could learn that words like man and woman correspond to the feature 21. Same would happen with more abstract terms, such as grammatical suffixes and such, as the meaning gets broken down further into fewer and fewer labels.

Anonymous at Thu, 23 May 2024 06:28:41 UTC No. 16189053

>>16189046
forgot pic

🗑️ Anonymous at Thu, 23 May 2024 06:50:23 UTC No. 16189072

>>16189053
Then you can be noticed how languages in populations with typically poor neocortexes contain very large number of premade words, while other only contain a limited number of morphemes, and a whole bunch is used at once in a huge "word" that corresponds to the features of what is meant to be described.

🗑️ Anonymous at Thu, 23 May 2024 06:51:24 UTC No. 16189073

>>16189053
Then it can be noticed how languages in populations with typically poor neocortexes contain very large number of premade words, while other only contain a limited number of morphemes, and a whole bunch is used at once in a huge "word" that corresponds to the features of what is meant to be described.

Anonymous at Thu, 23 May 2024 07:07:56 UTC No. 16189097

>>16189053
Then it can be noticed how languages in populations with typically poor neocortexes contain very large numbers of premade words, while other only contain a limited number of morphemes, and a whole bunch is used at once in a huge "word" that corresponds to the features of what is meant to be described.

Anonymous at Thu, 23 May 2024 07:11:34 UTC No. 16189101

>>16189097
This is why people complain about languages getting degraded, which linguists deny, as they only count the complexity of syntax, try to explain morphology as just a form of syntax, and don't count or notice this other process.

Anonymous at Thu, 23 May 2024 07:31:49 UTC No. 16189119

>>16189101
Note that I specifically don't call them "polysynthetic" as many supposedly isolating languages do it as well. (such as Chinese)

Anonymous at Thu, 23 May 2024 10:42:41 UTC No. 16189290

>>16188268
>society tries to turn us into controlled cyborgs through iconography.

Well yeah communication is important, building off already established ideas to innovate. Maths is much the same, you sacrifice a bit of your mind to dedicate it to an established collective to build off and utilize.

Anonymous at Thu, 23 May 2024 10:54:54 UTC No. 16189304

>>16188268
>Why does every emotion and thought need an external stimuli, it feels like society tries to turn us into controlled cyborgs through iconography.

More disturbingly still. All our iconography, grammar, symbolism, linguistics and etymology, are a response to the opioid receptors that dictate creative thought stifled after myelin sheaths grow into adulthood. They are the tombstones left behind by ideas bursting forth, and leaving them behind. The immune system of the body, like the mind, are one, and both as soon as they are born, are dying. It takes balls to look in the mirror and admit that. We. The walking corpse.

Anonymous at Thu, 23 May 2024 11:00:47 UTC No. 16189312

>>16189304
Then again. Is this not how we overcome all problems? Observation. The truth of the matter comes from the truth... of the matter. This is but an obstacle that once seen can be actuated upon. My understanding of math, chemistry and physics can only extend so far as my understanding, mastery and prowess of biology allows. So every so often. It good to draw a power from somewhere else.