These words are my own, from my “brain” flown

Our paper for this week presented an interesting argument- our way of speaking is in large part a result of innate processes. We as humans have a Universal Grammar, or “a facility for Language, ” that governs the way instantiations of Language (languages like English and Akan) manifest. There are parameters that we subconsciously set as we are exposed to language (like learning what order to put subjects, verbs and objects relative to each other in a sentence), but the parameters themselves are finite, and reflect our sense of Universal Grammar.

Why would we believe that a Universal Grammar, seeming similar in ways to Plato’s Theory of Forms, exists? Many things point us that way, including the fact that we as humans don’t ever get a large enough sample set to determine what sentences belongs in our language and what do not (it would be impossible to do this because there are an infinite number of sentences). However, from a young age we can pick out what sentences are correct and ill formed (even though we are often not taught which phrases are problematic with counterexamples.)  

The paper also postulates that our solid understanding of different parts of speech is not learned, but acquired. Though in grade school we are taught semantic definitions of parts of speech (we are told what a noun usually “means” in a sentence), we readily accept things that don’t fit into the semantic classifications as a member of that part of speech. The reason, our author postulates, is that we come to understand parts of speech as we acquire language. They are not consciously learned, but subconsciously acquired. For example, we estimate the probability with which words that end in “-ly” come between a determiner and a noun (in that case, zero) and use these internal distributions to form our notions of parts of speech.

This is where the argument breaks down for me. It was clear in the syntax case that we use probability to determine if a words is used correctly, or if it is a member of a certain class of speech (part of speech). As we are exposed to language, we develop working probalistic models to govern our expectations for language use. We would not argue that there is an innate sense of a particular part of speech, but that our understanding of that part of speech is developed with exposure (which is why small children will make ill formed sentences, but eventually grow to fix them). The same can be said for our use of language overall. Our author addressed the theory of statistical modeling governing our use of language briefly, but not robustly. Perhaps we “acquire” language by developing regular expressions based on the data we have gathered through experience, not because we innately anticipate them, but because we as humans are experts at seeing patterns in the noise of our utterances. We may not innately be prepared to learn grammar, but innately ready to find patterns. We understand ill-formed sentences in songs like Natasha Bedingfield's in this way, garnering meaning by making associations based on our previous knowledge:

Read some Byron, Shelly, and Keats
Recited it over a hip hop beat
I'm having trouble saying what I mean
With dead poets and drum machines

This verse is understood because of our acquired models for language and ability to find pattern. Is this a representation of an innate overall grammar? Don't you know, don't you know, I don't think so.