From Seeds to Syntax Trees

The Carnie reading covered many elements of syntax in a textbook-like manner.  The first part talked about how we as humans are preprogrammed to understand language through a universal grammar.  Nobody can teach us which sentences are grammatical and which ones aren't, because there are too many possible sentences in English alone—we just have to know by living in the world and observing how other people talk.  After approaching syntax from a philosophical stance, Carnie went on to define the different categories and subcategories of speech, most of which I knew but some of which were new to me (such as the determiner and the ditransitive verb).  I'd never really grappled with the problem of defining a part of speech before this reading, though I now realize the importance of both a morphological and a syntactical component of the definition.  With all the definitions in place, Carnie went on to discuss sentence diagramming with trees (or, less commonly, with brackets).

My initial reaction to the sentence diagramming discussion was, "Why, linguistics, why?"  I guess constructing sentence diagrams could be fun if you're into that sort of thing, but for me, it sounded like a real chore given all the complications.  I also didn't understand what the value of sentence diagramming was at first.  But everything started to click once the ambiguity of the "The man killed the king with the knife" example sunk in.  There really isn't any way to understand this sentence on a syntactic level without a sentence diagram, though one could technically deduce what's going on if given more context (which I suppose we'll learn about in semantics).  In any case, this reading reminded me a bit of last weeks readings (Gussenhoven and Kenstowicz) when taken as a whole.  Both the Gussenhoven reading and the first half or so of the Carnie reading presented us with many different ingredients (vocab words) in a linguistic recipe.  The recipe for last week's readings produced phonologic rules (as in the Kenstowicz reading), while the recipe for the first half of the Carnie reading produced sentence diagrams.  The cool thing about both these recipes is that they can help us document the current state of our language better for future generations and for artificial intelligence.

One thing that particularly interested me in the Carnie reading was the discussion of open vs. closed lexical subclasses.  I wondered whether any lexical subclass can be truly closed.  For example, English spoken 1000 years ago is very hard for a modern English speaker like me to understand, and it's not just because we have more instance of open class elements that we did 1000 years ago.  Carnie isn't necessarily wrong, but I feel he should've mentioned how members of closed classes can change over time and/or fall out of usage.  I also suspect there are periods of history where closed classes do undergo change—in this day and age, for example, I've noticed that the closed set of pronouns seems to be expanding in response to increased recognition of LGBTQ identities.  Thus, we can now refer to a gender fluid individual as "they" or a woman who identifies as transsexual as "he."