Nuances of language

          Carnie begins by disregarding the misconception that language can only be analyzed through a humanistic approach and applies the scientific method to it. Anaphors are the first noun that he scientifically analyzes: providing hypothesis, testing corpora, revising his hypothesis and continuing this process until, just like any other scientific experiment, there is enough well supported data to make conclusions. However, what is considered enough data? Carnie explains that the logical problem of language acquisition which shows that “productive systems are (possibly) unlearnable because you may never have enough input to be sure you have all the relevant facts”. I thought this was a very interesting point since it shows that even when you’re are extremely confident about a certain result or answer, it may not be correct, something that Carnie shows cleverly in his example of 1 to 1, 2 to 2, 3 to 3, 4 to 4, 5 to 5, and 6 to a number that breaks our expected pattern.

          Carnie then delves into the Parts of Speech topic where he in depth explain truly how complex language is. Depending on where nouns, verbs, adjectives and prepositions are located in a sentence, they can act as their one another. For example, assassination is an action, yet in the sentence “The assassination of the king” it is not considered a verb, but a noun. As Carnie points out, these complexities and nuances are found throughout all different languages. Not only that, but the meaning of the words can be irrelevant when recognizing parts of speech. We can read a sentence and without understanding the meaning of a single word, make out the verb from the noun and the adjective from the preposition.   

          The final chapter focuses on the structure of sentences, how to represent them in tree form, and the recursive nature of language, allowing us to create sentences we have never heard before. This chapter really made me think about the complex connection between linguistics and computer science. In theory it doesn’t seem very complicated to break down speech computationally since there are a limited number of words in language and there are patterns in connecting them; however due to the nuances of language that Carnie mentioned in the earlier chapters, it can become extremely tough to categorize and replicate what the human mind does when listening to sentences.  Yet in the past 10 years there have been significant advances in computational linguistics and word/sentence recognition. Siri, google search by voice and many other voice recognition software have all sprung up and it’s exciting to think about what the future holds for them. Could it be possible that intelligent computer operating systems like Samantha in the movie Her, become commonplace in households? That might seem unrealistic now however due to the speed of recent technological advancements, I feel that it could definitely be possible, if not probable.