From syntax to tree diagrams - the diversity of linguistics

Carnie begins his discussion with the assertion that syntax is a science. He states that sentences are generated subconsciously by a set of procedures, similar to computer programs, that form formal grammatical rules. We follow these rules subconsciously in order to form comprehendible sentences - we intuitively know how to properly order a sentence and punctuate it, step by step until it is a legitimate sentence. It is very interesting that we as humans can intuitively know that a sentence isn’t correct, yet it is difficult for us to describe the rules which are being broken, or the reasons why the sentence is incorrect.

I found the proof that Language is an innate, in-built system to be very interesting, because it relates to what we are learning in another class about finite state machines as models of language. This conception of sentence building makes the comparison to a computer program even more apparent - although a finite state machine cannot accurately or efficiently create all sentences, it can generate an impressive amount of language. The proof states that since syntax is a “productive, recursive and infinite system,” that it is therefore unlearnable. This is in line with the assertion that finite state machines cannot replicate language, since they run into the same problem - that language building is an infinite system that is recursive and therefore cannot be modeled finitely. 

Following this discussion of language generation as a science, Carnie describes how syntacticians follow a scientific procedure in observing language, making generalizations, developing a hypothesis, then testing that hypothesis against more data, and redoing the process if necessary. He then goes on to make a distinction between prescriptive rules of grammar and descriptive rules of grammar, the former being the kinds of rules that tell people how they should speak, and the later being the kinds of rules that describe how people actually speak.

In the second chapter, Carnie makes the interesting assertion that the part of speech of a word is determined by its place in the sentence and its affixes, not by its meaning. This means that we as English speakers can determine the parts of speech of various seemingly made up words of which we don’t know the meaning, for example “yinkish” (adjective), “dripner” (noun), “blorked” (verb) because of their morphology alone. 

In the final chapter, Carnie discusses structure and tree diagrams. The most interesting part to me was the variability in the tree diagrams of seemingly very similar sentences. The example given in the reading was the phrases “the big yellow book” and “the very yellow book.” The difference is found in the fact that “big” in the first phrase modifies “book” while “very” in the second phrase modifies “yellow.” This actually produces two completely different tree diagrams, and this illustrates how much tree diagrams can vary, if this slight modification of the sentence can produce a noticeable change. At the end, Carnie again brings up recursion and illustrates how language is recursive and that this fact allows you to create sentences that have never been heard before.