Anyway, this discussion is going nowhere. Notify me if there is other evidence apart from "I cannot imagine how the human brain can do that". As other people have written in a similar thread some months ago, this is pseudo-science.
From my own readings, I thought that there is very good evidence adults don't bother to correct children in a lot of cases, and that children ignore the corrections completely anyway. They also don't seem to make the kinds of mistakes that you'd expect.
A great summary of some of some research I've been reading on this is the paper [1]. Also Steven Pinker's early work [2] is fascinating and easy reading.
I would love to bring myself up to date though, if these things have changed lately. Any help appreciated.
[1] Gleitman & Newport paper The Invention of Language by Children - https://www.sas.upenn.edu/~gleitman/papers/Gleitman%20&%20Ne...
[2] Steven Pinker - Why the Child Holded the Baby Rabbits - http://socsci.uci.edu/~lpearl/courses/readings/Pinker1995Ch5...
Couldn't agree more. It's barely even that, it's kind of just armchair philosophizing. (which is fun and often actually useful, but not sufficient to make claims about universal laws)
All of this ends up with a huge combinatory explosion of possible ways to assign meaning to language, requiring a huge amount of examples and counter examples to actually learn a human language. Just look at the training corpuses of something like GPT-X to see the amount of information that is required to learn elements of a human langauge if you start without any assumptions.
However, we know for a fact that human children get many orders of magnitude less examples of languahe use before learning a language - far too little to actually rule out all other possible structures if they had started with 0 assumptions.
Adding to this, there are certain elements of language structure that are not commonly thought about that happen to work identically in all natural languages that have been studied, sucha s the fact that order of words in a sentence matters, or the way certain words are omitted when constructing embedded sentences (e.g. 'can eagles that fly swim?' always having 'can' refer to 'swim', never 'fly').