The structures I care about

I used to write about and teach literature. I read stories and poetry for a living. Why would anyone leave that?

Probably because I'm a fan of New Criticism. That's a problem because the current trends in poststructuralism and New Historicism are boring and they're powerful. So our break-up was mutual.

Where does a marooned formalist go? Why to linguistics of course.

In honor of National Poetry Month I'll revisit one of the old standards and say a tiny bit about that simple and much taught poem by the the good doctor William Carlos Williams.

        The Red Wheelbarrow
     so much depends
     upon

     a red wheel
     barrow

     glazed with rain
     water

     beside the white
     chickens.

Freshmen: He's just talking poetically because he thinks it sounds cool.
Sophomores: Each line looks like a wheelbarrow!
Juniors: Each line depends on one final word.
Seniors: He's a doctor and he feels powerless. The rain symbolizes his sadness!

Let's look at the Juniors' claim. Each line does depend on a final word. And each dependency is of a different sort. A quick summary:

First stanza: Once upon is uttered the poem must continue. And not because a preposition can't end a sentence. It's because the preposition doesn't yet have an object complement anywhere. The first line could be a complete sentence if we take "depends" in an absolute form. It's not fully standard in formal use but it is common. It's effective as a flat opening statement making a broad claim about life in general. The first line doesn't depend on the preposition upon but the rest of the poem does. Once we have the preposition the rest of the poem has to provide the noun complement or else it's not a full syntactic constituent.



Stanza two: This is a full constituent either way. The first line is just as grammatical with or without barrow. But once we know the intention of the poem (by reading ahead) we realize that there was a semantic dependency because wheel≠wheelbarrow. Only by knowing what the speaker means do we understand how the second line of the stanza is necessary because a word is incomplete. We might call this an accuracy dependency.



Stanza three: Again we have no grammatical dependency. We have also lost semantic dependency because rain and rainwater are pretty much synonymous. Williams has established a pattern by writing two stanzas with a similar form but different development in the structure of the two lines. But the structure of the third stanza mimics neither the developing structure of the first stanza nor the semantic rationale of the second. This is a stanza that follows form for the sake of form only. Now we know it's a poem for the sake of being a poem. It is my favourite stanza.



Stanza four: It's possible that white is a noun but it's not likely. It's reasonable to say that this is the only stanza in which the first line is unambiguously dependent on the second line. As the structure of the first line leaves out a necessary NP it is similar to the dependency introduced by the missing NP of the second line of the first stanza. So we have completion through a structural chiasmus: The first line of the last stanza at first sits as a non-constituent just while the last line of the first stanza becomes part of a non-constituent. And just as the first stanza is completed by the rest of the poem, the poem is completed by the rest of the last stanza.



I could say more about this but according to the rules I'm now required to discussed the hegemonic dynamics and the illusion of an original moment and the negotiation of cultural capital and the communities that will interpret these ambiguities and dependencies differently and what's a construct and what isn't.

I have more interesting things to do.

And I can barely make it in the prose

Poetry.com features a poem every day. Today's featured poem (for Saturday March 15, 2008) is Black Baldwin Grand by my friend Gretchen.

She's the saint who puts up with Casey's nonsense all day every day.

DS Eliot?

I don't feel like taking the time to go through and count every voiceless stop in the following reading by TS Eliot of his poem "The Lovesong of J. Alfred Prufrock". If I could I might figure out some sort of a pattern to his alternation of ±aspiration when /t/ /p/ and /k/ are word/syllable initial.



A quick listen reveals that the aspiration is a little stronger and more frequent on /k/. We hear "come" "coat" and "coller" pronounced [k^hʌm] and [k^həʊt] and [k^hɑlə]. Not heavy aspiration. He says "coffee" with much less. It's almost unaspirated.

There's also some evidence that when a /t/ is word final Eliot aspirates it. In the line "let us go and make our visit" the /t/ in "let" sounds slightly aspirated (there's not much flapping going on) and there's a clear aspirated release on "visit" at the end of the line instead of a glottal stop.

Of course there's also the phrase "of insidious intent to lead you" and the /t#/ of "intent" is clearly not aspirated -- which is expected because of the adjacent /#t/ of "to". It's the unaspirated word initial /t/ that sounds most conspicuous.

There's his pronunciation of "days" around 1:40. If an exceedingly intent speaker was to enunciate the consonants in "and days" it would likely come across as [ɛnd^h deɪz]. Normal speech would rarely separate the adjacent Ds. The pronunciation would typically delete the coda resulting in [ɛndeɪz]. Following "and" it's odd to hear a stop of voice. And that's probably what makes his pronunciation sound like [teɪz] to me: I don't expect to hear the break between an alveolar nasal and a voiced alveolar stop so I perceive the -voice as a feature of the stop.

Of course I wonder how specifically affected Eliot's phonology is. Pay attention around 3:00 when Eliot says "to spit out all the butt ends of my days and ways." The reading of a poem is a performance and it's not necessarily relevant to any questions of dialect or production. It's more relevant to questions of perception. The distinction between the aspirated and unaspirated voiceless stop onset is difficult to illustrate to students. I use a [t^h] and they recognize it easily. I get rid of the aspiration and they think it's a [d]. I alternate between [t⁼] and [d] as a minimal pair and they often can't even hear the difference. (Intervocalically it's easier.)

From now on I'll just play this clip for them. If they can't hear that it sounds odd I'll just give up and move on to the next topic.

It's a boy or a girl

I had a chance recently to go back and look over the Beowulf translation I completed a year ago. On the second day of class I was given the five or so lines ending with the memorable exclamation regarding Scyld Sheafson: "Þæt wæs god cyning" -- translated by Seamus heaney as "That was one good king."

When I got to the line I read it as "That's a good king" -- the professor just looked up at me and said "Well that doesn't help us out any." *sigh*

His contention was that because Þæt is neuter and cyning is masculine it just doesn't make sense grammatically to say that they were both referring to the same good king. His suggested translation? Something along the lines of "Thusly (he) was a good king." "That" had to serve an adverbial function not a nominative -- perhaps like that or in that way.

It made sense so I lowered my head with a quiet yessir and ceded to floor to the next trembling classmate.

Looking at my translation again it occurs to me that my first attempt might work. Think of "That's a good king" as All that behavior adds up to a good king.

Or let's look at another language.

I say of my wife 'That is my wife' and I repeat it in Spanish: 'Ésa es mi esposa.' The demonstrative agrees with the gender of the predicate nominative.

And if I use the definite article saying 'That is the best wife' my Spanish echo would say 'Ésa es la mejor esposa.'

But if I change the sentiment slightly and say 'That is a good wife' I would say it in Spanish: 'Eso es una buena esposa.' The determiner on the nominative NP constrains the demonstrative.

A possessive forces agreement:
Ésa es mi amiga.
Ése es mi padre.
Ésa es mi casa.

(I have a sense that perhaps a flippant phrase like 'that's my friend' delivered with a sigh meaning 'what more can we expect from her/him?' could take the neuter. I'll have to look into that.)

But an indefinite article is fine with the neuter demonstrative.
Eso es un padre.
Eso es una casa.
Eso es una mujer.

If we use an intensifying quantifier in English: 'That is one big house' -- the Spanish counter would be 'Ésa es una casa grande' -- The quantifier looks identical to the determiner but each functions with a distinct demonstrative.

English: That is a big dog.
Spanish: Eso es un perro grande.
English: That is one big dog.
Spanish: Ése es un perro grande.

Even tho the article forms are identical in Spanish the sharper semantic focus is evident from the demonstrative. Perhaps one≠one. (This may be further evidence that dividing determiners and quantifiers into definite and indefinite determiners is problematic.)

For the sake of analogy (but not argument) I suggest that similar constraints on agreement might work in OE. When the demonstrative might indicate an collection of behaviors or qualities more than an individual or an item a neuter can function even when equated with a masculine or feminine noun.

The line in Beowulf doesn't have an article. OE didn't have true articles. But because we have neuter þæt and not masculine se a translation into Modern English could easily be That was a good king distinct from He was a good king. If we accept this analogy we then have to reject Heaney's translation: "That was one good king."

But I'm loath to criticise any decision Seamus Heaney makes regarding poetry. He's earned his license.

(Yes--the line from the MS was cropped and reorganized to fit the space.)

Nabokov Schmabokov

It's 'naBOKov'

John Wells has given us some entertaining guidance regarding the pronunciation of Vladimir Nabokov (i.e. the pronunciation of his name/s of course). I grew up thinking it was a dactylic dimeter. And by "grew up" I mean from the time I first heard the name in junior high from my sister who loved Sting and decided to learn more about Nabokov after hearing his name in the song "Don't Stand So Close to Me."

3 poems to make the point that each name is not a dactyl. Lines 1-3 of each stanza are dactylic dimeter. One line has to be a single word. Lines 4 and 8 must rhyme. (These were the regulations of the New Statesman competition. The third poem below was the winning entry.

Humberty-nymphety,
VN the novelist
Wrote of a passion pro-
scribed by the law.

Careful! His names must be
Stressed à la russe, i.e.
Paroxytonically:
("...dímir Nabó...").
-Nigel Greenwood

The next one is in Dutch. The rhythm is the same so you should be able to "utter" it and hear how it works. I'm assuming the "Na-" of "Nabokov" in the final line is orphaned from the second dactyl in the preceding line. Translation follows:

Vlinder- en kindervriend
Vladimir Nabokov...
Lezer, uw uitspraak doet
Pijn aan mijn oor.

Volgens de Russische
Intelligentsia
Moet het Vladimir
Nabokov zijn hoor!

Translation:
Lover of butterflies and children,
'Vladimir 'Nabokov...
Reader, your pronunciation
hurts my ear.
According to the Russian
intelligentsia,
it ought to be Vla'dimir Na'bokov!
-Jan Kal

The winning entry:

Higgledy piggledy
Vladimir Nabokov —
Wait! Hasn't somebody
made a mistake?

Out of such errors, Vla-
dimir Nabokov would
sesquipedalian
paragraphs make.

Why should we care? Only because Nabokov himself cared. Wells writes: "Apparently Nabokov himself characteristically liked to point out that his first name rhymed with Redeemer."