Featured Post

BFRC

I am posting this as a benchmark, not because I think I'm playing very well yet.  The idea would be post a video every month for a ye...

Saturday, May 22, 2010

Scientific Method

I have a research problem that, improbably enough, is susceptible to empirical information. What I want to know is this: how perceptible are line-endings (and by extension the identity of lines themselves) in regular and somewhat irregular Spanish verse,, whether rhymed, assonantally rhymed, or unrhymed? How perceptible is rhyme herself and how does the presence, absence of rhyme affect this? Obviously, line endings are very perceptible when you look at them on the page. I mean when someone is hearing authentic performances without looking at a text.

So obviously I need the following: enough authentic performances of enough poets. By authentic I mean here nothing too profound. An authentic performance is one by the poet him or herself, or one arising out of the same cultural context of the poet. What I don't want to do, for example, is have the research subjects listen to me reading the texts out loud.

Secondly, research subjects, human beings willing to participate. i need at least two groups, one more expert than the other.

Thirdly, I need a protocol. What are these subjects going to do to demonstrate their perception or non-perception of line-endings? Take a test or questionnaire? How should this be designed to really answer the question I am posing?

I need a method of interpreting the data, making sense of what I have found so that it will be valid and meaningful. I need to know what to do if the data pull strongly in one direction or the other. I need to find a way of defeating my own biases.

I also need a rationale for why I am doing the study. What is the question behind the question? What are the implications, the consequences? So what?

The funny thing here is that, though I haven't ever done anything like this, and don't know how in technical terms, I at least know some of the elements that would go into a study like this. I would be suspicious of my own ability to make it scientifically valid.

5 comments:

Thomas said...

One worry I would have is how you determine that the line in fact ends, which must obviously be done if we are to decide whether people are able to perceive that ending. My feeling is that the way poems are presented on the page don't always get "the identity of the line" right.

Why not just give research subjects a poem with the line endings removed (shaped like a paragraph), and ask them to lineate? Great agreement among subjects would show that line endings are easy to spot.

The good thing about this is that agreement would be independent of the poet's own sense (perception) of the line endings. If 80% of the respondents agree with each other, but only 5% lineate like the poet (with the remaining 15% lineating all over the place), then it says that the poet may write a good line but doesn't know how to lineate (i.e., according to intersubjective standards of "the identity of the line").

Interesting, enjambment might not always be "perceptible" except in performance (or of course in the poem as a written). This effect sometimes breaks against "natural" lineation.

My feeling is that most many line endings are (too) clearly marked in perfomance, except when the reader makes a mistake, or when the line ending in the poem as written was arbitrary.

I don't know much about spanish poetry, but I have feeling that there would be great agreement among readers in lineating WCW ... interestingly, I'm not sure that agreement would necessarily "line up" with WCW's own lineation. But there is a very natural rhythm in his verse, which I think readers would parse out in roughly similar ways if they were asked to put in the line breaks. I think his line is "perceptible", even if we were forced conclude that he was effectively trying to hide the endings on the page. (I'm not saying that as a hypothesis, though; I haven't thought very much about this ... yet).

PS. I love how you say "rhyme herself" not "rhyme itself". Is that conventional in Spanish lit?

Jonathan said...

Thanks for the suggestion. Don't you think readers lineating a poem on the page would count the syllables and look at the vowels to find the rhyme? Spanish poetry, even free verse, tends to still divide up into classical lines more or less. Sevens, elevens, and fourteens with some nines thrown in the mix.

I'm thinking of using your suggestion as a control. In other words, another skill apart from what I am looking at. It would be interesting if students could do something on the page they couldn't do by listening--or vice versa.

"rhyme herself" was just my jocularity, not standard practice. The word rhyme is feminine in Spanish.

Thomas said...

Well, perhaps syllables are what makes line endings visible. My counter-worry is that the performed poem would "telegraph" the answer, which wouldn't tell you what you want to know. (If they simply mark the line where the poet inhales, for example.) Maybe I don't have the right image of an "authentic reading" in mind.

Jonathan said...

Preliminary tests show that this telegraphing does not occur. Obviously some lines end with the end of the intonational phrase, but when they don't, then ambiguity ensues. I want to see how rhyme or its absence makes line endings more or less perceptible in the auditory perception of poetry.

Vance Maverick said...

I think talking to a statistician would be illuminating. Some, of course, are focused on mathematical sophistication, but many are interested more in how to frame and answer questions about the real world. And there are others at the U (psychoacousticians, experimental psychologists) who might have ideas for how your subjects should respond, and even an experimental rig you could use. (For example, you might have them tap while listening to the recording -- then look at where the taps fell w.r.t. the verse.)