One thing leads to another. First I ran across linguist and cognitive scientist D. Girard Watson, and his excellent piece on those speech fillers we all use, the Ums and Ers.
Then I sent the link off to D2 (daughter #2; she prefers to remain anonymous on my blog), because she counts linguistic and neuroscience studies among her varied pursuits, and it kicked off a conversation. Below I post her comments in interview form.
Me: So, what did you think of Watson's piece?
D2: Great article for comparing written and oral language. We glean so much from speech -- cues that are embedded in emphasis and tone, speech rate, word fillers, and cues from dialogue (like being interrupted or "latching" on to speech to finish another person's sentence). Those kinds of cues are not just about understanding one another's speech -- they are cues that convey information about the relative stances or positions of power in a dialogue, etc.
Me: So, these verbal cues tell us which speaker is the Alpha and which is the Beta dog?
D2:In fact, most speech I hear is "doing" something rather than saying something-- that is to say, speech that does not primarily exist to convey a literal message, but is used in its context to reproduce relationships of power, e.g. reinforcing a person's authority (the way a professor answers a question by avoiding it entirely; in academic settings there are endless examples of "displays of knowledge"), conveying solidarity between friends ("You know what I mean?" "Yeah I got you"), etc.
Me: OK, it's easy to get authority or solidarity across in spoken language, especially because we can use tone and body language to accompany spoken words. How do we do this in the written (novel) form?
D2: To a certain extent I think that speech variation can be portrayed in novel writing, but there are limitations because of the form. Dialect and slang are a couple examples. Writing and grammar are governed by prescriptive rules that are more resistant to change, whereas extemporaneous speech is not (and so sociolinguists are interested in descriptive research rather than prescriptive norms). In writing, for example, an omnipotent, third-person narrator does not usually use dialect or slang: Because the narrator is anonymous, he/she tends to assume a grammatically normative style. That's not to say that all narrative voices sound the same -- unique narrative styles distinguish one author from another -- but they are usually the most "proper" voice in the novel.
There are a lot of places you could go with this discussion, but I think another point about normative grammar is that it is fulfilling the reader's expectation for an "objective" narrator. Authors sometimes bypass this by narrating through a known character, often a bystander or protagonist telling their story in hindsight a la Moby Dick or Heart of Darkness.
Me: Yes, even in first person POV the narrator creates a "normative" style. I don't write in first person, but I get it.
D2: Another point I wanted to make -- a narrator's voice may be marked by the style and language of the period, but only insofar as the narrator speaks in a socially "credible" (educated, upper class) language. Dialects that are not perceived as credible or standard (a Cockney accent, a foreigner's pattern of speech) are given voice through character dialogue or through a first person narrator. In that way, historical novels are documents of the sociolinguistic landscape of the time -- not just types of language, but their ranking in the social order.
Me: Thanks, D2! Um, have I told you how much I, um, love you?