Tuesday, July 11, 2006


When my father, a writer and academic, died about twelve years ago, I inherited a lot of books and papers. Yesterday I went through them to clear some space. While sorting out a long row of literary magazines to keep the ones he'd edited or published something in, I ran across an issue of The Green Caldron: A Magazine of Freshman Writing from the University of Illinois, December 1956. My father began there as an undergraduate just before he went into the war, and finished in the late forties just before I was born. Though I might not have seen it otherwise, in the close search for his name in the tables of contents (probably closer than it would take for my own name to jump out at me), I noticed a piece — "My Theory of Religion," apparently a set topic in Rhetoric — by Dennis Jay Zeitlin.

I've admired the jazz piano of Denny Zeitlin, who is also San Francisco psychiatrist, ever since what may have been his first recording session, on Jeremy Steig's Flute Fever. That LP has never been released as a CD; after many tries I got a copy on eBay last year. Steig's playing on his debut album is as great as I had remembered, and so is Zeitlin's: "accompaniment" doesn't give quite the right picture, and a couple of his introductions, as well as his oblique solo on "What Is This Thing Called Love," stuck with me well enough that during the decades while I didn't have the album I could replay an internal tape of those bits in my head. The web, bless it, confirmed Zeitlin's full name, gave me a birth date that matched, and even furnished an e-mail address, so I was able to write and ask him if he'd like to have the magazine.

What does this mean? Coincidences are meaningless, more or less by definition (otherwise we go all far-look-in-the-eyes and call them syncrhonicities), but confluences aren't. We seem to differentiate events that merely take place, so that if they take the same place they coincide, from events that constitute a flow, or more than one, so there can be a conflux of them. The things flowing together here flow only in my own mind's life. Also in Zeitlin's, but for him it's a moment's return to an earlier point in the main stream. For me there's the Father stream and the Jazz stream. It's not just that the magazine with Zeitlin's freshman essay turned up in my father's books (oddly, since my father had graduated a few years before). I heard Flute Fever as early as I did — it helped form my idea of jazz along with Kind of Blue, which I heard at the same time in the same way — because somebody gave my father a copy of the album on a reel-to-reel tape. I remember listening to it with him.

The topic of Zeitlin's freshman essay doesn't happen to enter into the confluence, for me, and it can only be in me that we're talking about here. If it did, the sense of a meaning would increase, both because there would be some third stream in the mix, and because there would be substance, something to substantiate the significant form of confluence.

Thursday, July 28, 2005

Beyond Acuity

Meanwhile: The parsers published by the Stanford Natural Language Processing Group (here) go a good distance toward providing the automated parsing needed to calculate "acuity" programmatically. Not all the way, since they can be confused by complicated sentences, and especially ones whose syntax is colloquial -- just the kind of sentences poems tend to be full of.

The more I experiment with this, however, the more I realize it's only a partial approach to specifying an enjambment's force in a poem. It still doesn't, and possibly can't, determine what I called "potential acuity." The Stanford parsers insist, all too successfuly, on treating whatever string-of-words they're given as a grammatical sentence. The question of whether a line could end a sentence -- which I still think is a vital component of how we read the line-break -- turns out to be more difficult than I realized. The second line of Wordsworth's Westminster Bridge sonnet, for example: "Dull would he be of soul who could pass by". You know and I know that "pass by" is going to be transitive, and putting a period after it would feel unnatural. But "pass by" can work like its synonym in the song-title "Walk On By"; nothing in the syntax seems to support the instinct that the line must be incomplete.

Furthermore, syntax is only part of the story. (See Ray Jackendoff's Foundations of Language for a related argument against "syntactocentrism" in linguistics.) For example, we would also want to know whether an enjambment occurs at the end of, or in the midst of, units that are prosodic in linguistics' sense of the word, such as the Phonological Phrase, Intonational Phrase, and Phonological Utterance. Those appear to be more fungible than syntactic units, raising the additional possibility that a line-break will create a break between prosodic units.

Enjambment begins to look like an arena in which many kinds of linguistic processes and structures are brought into play, or in which the play within and among them is brought into prominence and the reader's attention. On the bright side, this might make enjambment a good, rich test-bed for the study of multiple "interface modules" of the kind Jackendoff describes.

Sigh. Such as it is, my public career as a literary critic began back in 1981 with a book (Free Verse: An Essay on Prosody) that concentrated on line-breaks as "prosodic" (in the literary sense: devices for governing rhythm). Now I'm back to the same topic about 137 years later. I'm even re-fascinated by one of the main examples I studied back then, Ashbery's "Self-Portrait in a Convex Mirror." Now supplemented, I'm glad to say, with other examples to turn over and over in the mind, such as Phil Levine's delightful poem "A Theory of Prosody."

Sunday, May 01, 2005

Acuity: A Metric for Enjambment

We know enjambment is an important means of rhythmic control in all verse, and vital in free verse; we know that enjambments have different "strengths"; but we have no definite way to measure and compare these strengths. I propose a measure tentatively called "acuity." ("Gradience"? One emphasizes the poet's act of cutting into the sentence with more or less vigor; the other emphasizes the reader's experience in traversing the enjambment.)

Determining acuity begins with a complete parsing of the sentence that contains the line-break. The tree diagrams, or equivalent labeled-bracketings, common in linguistic studies of syntax, provide a convenient basis. Here's an example from Radford, Transformational Grammar, pp. 53-54: "This boy must seem incredibly stupid to that girl."

[S [NP [D this] [N boy] ] [M must] [VP [V seem] [AP [ADV incredibly]
[A stupid] ] [PP [P to] [NP [D that] [N girl] ] ] ] ]

The words of this sentence, and the acuity of a possible line-break following each word:

this 2
boy 1
must 1
seem 2
incredibly 3
stupid 2
to 3
that 4
girl 0

To measure the acuity of a line-break following a word, we count opening brackets to the left of the word, subtract closing brackets to its left, and subtract the one or more closing brackets to its right preceding the next opening bracket (or non-bracket, non-punctuation text) if any. For "to" in the sample sentence the measure is 12 - 8 - 1 = 3; for "boy" it is 4 - 1 - 2 = 1; for "girl" it is 15 - 10 - 5 = 0. The proviso about non-bracket, non-punctuation text means that a line-break in the middle of a word ["in- / credibly"] has an acuity higher by one than a line-break at the end of the word.

Among many practical difficulties, one complication is that syntactical ambiguity, which is common, may allow the sentence to be parsed in more than a single way.

Since this metric cries out for a computer program that would catalogue the acuities of a poem's enjambments, a vital practical question is whether the state of the art has arrived at a point where completely parsing arbitrarily complex English sentence is possible. I don't know the answer; I'm trying to find out.

The next step is to recognize potential acuity. In the formulation above, the crucial qualification "uncompleted" suggests a separate characterization of enjambments based on whether each node to which the pre-line-break word belongs might or could be complete without damaging grammaticality. The relevance of this addition shows up most sharply in possible enjambments of "garden-path sentences":

The horse raced past the barn

The metric proposed so far assigns to the line-break after "barn" an acuity of 1. Intuitively speaking, it should be much higher, because the first line can so readily be parsed as complete. In general, the possible unfoldings of a sentence -- as opposed to and preceding the actual -- influence the felt sharpness of the line-break. When a poem begins with the line, "This is the time of year" (Bishop, "The Armadillo"), a potential acuity of 0 vies with what turns out to be the realized acuity of 1 or 2. In the sample sentence above, the same would be true of a break after "stupid."

Is it true, as it seems, that potential acuity can never be greater than realized acuity? It is not true semantically or pragmatically; the sentence's continuation can constitute less of a surprise than seemed promised. (This renders the line-break less interesting, and poets therefore avoid it. This makes examples difficult to find, though not to construct.) The question about measures of acuity is whether a similar syntactical anticlimax is possible.

As a first, sentence-level measure of the potential acuity of a line-break, we could hypothesize a period at the end of the line and seek a complete parse of the sentence as if it ended there. Extensions to lower levels of the parse tree might be complicated but should not require breaking different conceptual ground.

Potential acuity, however, is inevitably more difficult to specify deterministically than actual acuity. When the sentence and its complete (and unique) parse are given, actual acuities can be derived by simple counting of brackets. Hypothetical sentence terminations raise questions that are not strictly syntactical; in "This is the time of year," the semantics of "the" increases the probability that the sentence will continue so as to specify which time of year the sentence is undertaking to designate.

Scandroid 1.1

Version 1.0 of the Scandroid was as major upgrade. The program now handles anapestic as well as iambic meters. Now a version 1.1 is available, including mostly internal improvements for more accurate performance. The link given in the previous post still takes you to the right place.

Whether the Mac version will work correctly with Tiger (OS 10.4) I'm not entirely sure, but I'll know (and begin working on any necessary adjustments) by the end of the week.

Monday, February 28, 2005

the Scandroid Arrives

It's a program that scans iambic pentameters. It does a good enough job that its mistakes are interesting. It's available for Mac (OS 10.3 only) and Windows:

Download the Scandroid

The Manual is also available there as a separately downloadable PDF file (it's included in the Mac and Win downloads too).

This is version 0.2a. Version 1.0 is in the works, and it offers important additions and improvements. Most important, it will handle other meters, anapestic ones in particular. Figuring out how to decide — how we decide — whether a line is anapestic or iambic is difficult enough to constitute a reason for writing the program. I'm collecting other questions that the program raises. If anyone finds others, and of course if anyone finds bugs, I'd be very glad to hear.

The program is published under a GNU Public License. The source code (in Python) is available on the same site as the program itself.

Thursday, February 24, 2005

Announcing: the Scandroid

It's a program that scans iambic pentameters. It does a good enough job that its mistakes are interesting. It's available for Mac (OS 10.3 only) and Windows: http://cherry.conncoll.edu/cohar/COH%20programs%20page.htm. The Manual is also available there as a separately downloadable PDF file (it's included in the Mac and Win downloads too).

This is version 0.2a. Version 1.0 is in the works, and it offers important additions and improvements. Most important, it will handle other meters, anapestic ones in particular. Figuring out how to decide — how we decide — whether a line is anapestic or iambic is difficult enough to constitute a reason for writing the program. I'm collecting other questions that the program raises. If anyone finds others, and of course if anyone finds bugs, I'd be very glad to hear.

The program is published under a GNU Public License. The source code (in Python) is available on the same site as the program itself.

Thursday, February 17, 2005

        Hole Up

February's made
me think
on it, why
out, or in of
course, or
down, at least.
Myself am
hole, nor am I
out of it.

Tuesday, February 08, 2005

Politics of the Random

The Sinking Ark (2)

Steven Levy, an editor at Newsweek, gets an iPod Shuffle and finds that it keeps playing his Steely Dan cuts in preference to everything else he's put on it. Being a Newsweek editor with Clout, he calls Steve Jobs, who tells him that the random-choice algorithm has been rechecked and validated. Being no dummy, Levy next calls a mathematician.

The mathematician lets Levy in on an open secret: our brains can't do random, and what's more, won't do it. They (which means we) would rather posit a pro-Steely-Dan conspiracy among the chips in the iPod or back at the Apple factory or in the basement of the White House than acknowledge that a genuinely random sequence usually seems to us full of pattern and meaning.

Michael Gazzaniga, who spread the word about split-brain research back in the seventies, published an article in the July 1998 Scientific American called "The Split Brain Revisited." He recounts a wonderful experiment at Dartmouth in which commissurotomy patients' separate brain hemispheres are each presented with a weighted random stimulus: two lights, one of which flashes 80% of the time and the other 20%. The right hemisphere quickly settles down to gaily guessing top top top top top, and a success rate of 80%. So does a rat. The left hemisphere won't quit; it clings tenaciously to the idea of a pattern, constructs elaborate systems of prediction, and never gets above 68%. It knows somebody in there is pushing Steely Dan. From the left hemisphere's point of view, the right hemisphere's behavior is mindless.

The most vicious and stupid manifestation of the Steely Dan Maneuver, or the stupidest and most vicious on display this morning, is the campaign to teach "Intelligent Design" alongside the stuff Darwin pointed out 150 years ago. The fundamental argument, of course, is that randomness just couldn't produce something as complicated as life, the eye of the octopus, Steely Dan, or the Kansas Board of Education. My own hope is that Kansas will enact this "educational" policy, because I'm in favor of intellectual and financial resources moving from the middle of the U.S. toward its edges—as Garrison Keillor remarks, state lotteries are a tax on people who weren't that good at math—but I realize this is provincial, so I regret hoping it.

The best thing about the SDM, on the other hand, is described this way by Gazzaniga: "Our uniquely human skills may well be produced by minute and circumscribed neuronal networks. And yet our highly modularized brain generates the feeling in all of us that we are integrated and unified. How so, given that we are a collection of specialized modules? The answer may be that the left hemisphere seeks explanations for why events occur. . . ." In a word, that rabid left hemisphere fabricates our sense of having a self. (And this is the best of the SDM?) Theories of God couldn't have been far behind, a few decamillennia ago. That would make the Kansas argument well and truly and cosmically circular.

Saturday, February 05, 2005

Watch This Space

Most of my past month (between semesters) has gone into work on the Scandroid, a program (Mac and Windows) that scans iambic pentameter. A usable first version, though still very much alpha stage, is almost ready. I'll post links for the executable and source code when they're ready; and a series of notes on the implications of the program and its methods would make as much sense here as anywhere.

It's not the first time I've written such a program. My Virtual Muse: Experiments in Computer Poetry, a book that's been out of print for a few years, mentions in passing a very early version of "the Scansion Machine." This one builds on that, but I goes farther, does better, and at least in my thinking about it, goes deeper. Incidentally it is as far as I know the most successful program around at the specialized thing it does: producing traditional-looking scansions of English metrical verse.

More soon! Stayed tuned! Alert the press & don't let the kids go to bed!

Monday, January 03, 2005

Robert Matsui (D-Ca) died yesterday, and the radio used the same phrase as always. Why is it "Japanese internment camp" but "German concentration camp"?

Friday, December 24, 2004

To What Purpose

Except to Be (3)

Though I respect the intensity of my students who feel otherwise, I can't much admire the poems of Edna St. Vincent Millay, which seem to me more serious about the poet's pose in the world than about either the world or herself. Her "Spring" — "To what purpose, April, do you return again?" — says: You won't catch me getting taken in by this rebirth crap, knowing as I do, from my history of hurt, how . . . etc. "I know what I know," one line congratulates itself.

Yet that famous opening line holds something fundamental to poetry, not because of its attitude or its attitudinizing, but because of its absurdity. The ideal Logical Positivist — indispensible joker — will object that, even given the poetic fiction of seeming to address a segment of the calendar, questioning its intent and motives misapprehends such fundamental distinctions as the one between sentience and insentience, and above all the incomparability of time (which goes by, one hour per hour, no matter what you think or want) and creatures in time, which go on as they can, until they don't. The poem begins from, and can't do without, and replaces at the center of our consciousness that created it, an opposite conviction. If we can't question April, what can we question? If we can't question April, who can?

So I don't dismiss Millay's premise. I miss her taking it more thoroughly on its own terms. To do that would have led her into myth. The rest of the poem would wonder through that myth, not keep distracting itself and us with veiled references to past disillusionments. She ends by comparing April with Ophelia — if only she had gotten there first and gone on, not saved it for the end. To April's Ophelia, whose Laertes? Polonius? Hamlet? Could the poet have ended up as Gertrude instead, wary and pitying and doomed?

Friday, December 17, 2004

Fake Book

Vill Ex (2)

Few people in the general public — say, the roomful at this moderately upscale soiree — seem to notice that the jazz group in the corner represents a minor miracle. It came to be here perhaps by way of the hostess’ niece’s phone call to her friend the bass player, who asked around and got numbers for a pianist, a drummer, and an alto saxophonist. The musicians have never met before. This should be a problem. A smooth, finished sound is even more important to functional background music than to explicitly adventurous concert performance. That smoothness demands what we loosely call “professionalism,” but at least it seems logically to require a certain amount of rehearsal.

What allows them to produce the effect without its apparently necessary precondition is a fake book, which is a collection of “standards” — which here simply means tunes or kinds of tunes the musicians think the partygoers want or expect to hear. The fake book displays one tune per page: not representing it, as in published sheet music, by a vocal part accompanied by a piano reduction of some real or notional orchestral arrangement; epitomizing it, instead, in a single line of melody with a sequence of chord changes written above it.

The Oxford English Dictionary seems not to have heard of the phrase fake book.* In this context, “fake” isn’t the adjective people might take it for, meaning “ersatz” or “imitation,” as in “fake cream” or “a fake Vermeer”; no original object exists which the collection quite copies. Rather, “fake” is a verb, and a fake book is a book which allows the musicians to fake it. What they are “faking” is a question which the OED evades by means of a special definition for this sense of the verb: “Of jazz musisicans: to improvise. Colloq.” Are they (to seek help from other definitions) “feigning or simulating” something? They are a jazz group, no matter how new and transitory, not a simulaton of one. Are they “concealing the defects” of something, like a horse one wants to sell? Possibly, though if the group is so superlative that there are no noticeable defects, the term ought then to lose its force, but does not.

“Fake” as a name for the defining activity of jazz musicians has been part of English at least since 1926. OED’s first exemplary quotation refers specifically to the situation in which there is a written but incomplete score, from which certain instrumental parts must be inferred without benefit of notation. To fake is to extrapolate. By the time of the next quotation (1933), the reference seems to be to much more pervasive structural improvisation, on the standard jazz model in which the written framework of performance is confined to that single melodic line with harmonic progression which eventually gets collected in fake books.

Back then, the jazz group often was treated as a mimicry: it imitated a non-jazz band, one that performed “hot” music from a written score, like Paul Whiteman’s. This wry attitude toward their own art characterized jazz musicians themselves for decades; thirties and forties players would speak of their individualized instrumental voices as deviating from “legitimate” tone, contrast their work with “serious music,” and call the group’s intonation “close enough for jazz.” All this changed when the music began to take itself seriously, a process that began with bebop and reached a kind of apotheosis in John Coltrane’s dedication to his art in the late sixties.

Now, of all the fake books the jazz world has seen since between the World Wars, none has had the longevity or primacy of The Real Book. Delicious title. We don’t know — I don’t know — who made it or who still occasionally distributes it. It seems to have been first put together in the mid-seventies. It’s a curious cultural artifact: on the one hand blatantly illegal (no royalties have been paid for the tunes), sold out of the trunks of cars; on the other hand, essential for any working jazz musician, who is expected to “know The Real Book” as a pre-requisite for work in any pickup group.

Like a Rolex in reverse, The Real Book is so meritorious that legal imitations of it abound. A Dadaist could photocopy one of these, then point to the copy sitting on a music stand and, to distinguish it from a photographic or memory image of it, call it a real fake fake Real fake book.

*Not true, as Forthright (http://phrontistery.info) points out; but OED puts the phrase under the noun "fake," not the verb — wrongly, I think.

Tuesday, November 23, 2004

Sheep, Goats

Vill Ex (1)

At the beginning of this year there were six million sheep in America. The decline has been almost steady since the peak year, which was 1942, when there were fifty-six million. (Almost steady; there was an odd upward glitch at the end of the fifties.) There are a few more than a million goats.

"Sheep" and "goats" usually occur together in Judaeo-Christian scripture as an equal, uninvidious choice of victims: "if his offering be of the flocks, namely, of the sheep, or of the goats, for a burnt sacrifice; he shall bring it a male without blemish" (Leviticus 1:10), etc. Our common phrase is an innovation introduced in Matthew 25:31-32. It doesn't come up in Revelation, but sure prefigures it: the Son of Man, enthroned at last, with the nations gathered before him, "shall separate them one from another, as a shepherd divideth his sheep from the goats: And he shall set the sheep on his right hand, but the goats on the left."

Christianity, though, probably didn't infuse judgment into the distinction out of thin air; there's that matter of the scapegoat, which comes from fully as early as the take-your-sheep-or-goat-to-the-altar passage in Leviticus. Just two verses earlier, Aaron is told to "cast lots upon the two goats; one lot for the LORD, and the other lot for the scapegoat." From the goat's point of view, this counts as a Trick Question. The Lord's goat gets offered "as a sin offering," which presumably entails some throat-slitting and selective incineration. The other "shall be presented alive before the LORD, to make an atonement with him, and to let him go for a scapegoat into the wilderness." Ah rapture!

As a practical matter, "when you can't tell the sheep from the goats by their coat," a helpful website advises*, since Angora Goats are pretty woolly, "you can look at their tails. Sheep generally carry their tails hanging down, and don't tend to wriggle them very often while goats have ver[y] mobile tails which they often hold erect." Just so.

* http://www.fortunecity.com/marina/bounty/170/goats.html