Language Log

« previous post | next post »

Unexpectedly, Stanley Fish's most recent NYT essay praises word-counting as a technique of rhetorical analysis ("Barack Obama's Prose Style"):

One day after the occasion, USA Today offered as an analysis of [Obama's inaugural address] a list of the words most frequently used, words like America, common, generation, nation, people, today, world. This is exactly the right kind of analysis to perform, for it identifies the location of the speech’s energy in the repetition of key words and the associations forged among them by virtue of that repetition.

This is odd, since Prof. Fish is known for attacking attempts to base literary analysis on counting things in texts (e.g. "What is stylistics and why are they saying such terrible things about it?", in Essays in Modern Stylistics, 1981). But he returns to type in his own analysis of the inaugural address, which relies on quantitatively unsupported assertions:

There are few transitions and those there are – “for,” “nor,” “as for,” “so,” “and so” – seem just stuck in, providing a pause, not a marker of logical progression. Obama doesn’t deposit us at a location he has in mind from the beginning; he carries us from meditative bead to meditative bead, and invites us to contemplate. […]

There is a technical term for this kind of writing – parataxis, defined by the Oxford English Dictionary as “the placing of propositions or clauses one after the other without indicating . . . the relation of co-ordination or subordination between them.”

The opposite of parataxis is hypotaxis, the marking of relations between propositions and clauses by connectives that point backward or forward.

Ironically, this analysis is quantitatively doubtful, as we'll see below: Obama's speech actually uses the relevant sort of connectives at a higher rate than George W. Bush's 2005 inaugural address did.

Fish's analysis is also doubtful as a description of readers' responses: even in the paratactic parts of Obama's speech, it's clear that the reader is generally "deposited" at a series of discursive "locations" that are at least as carefully calculated and logically related as in, say, George W. Bush's inaugural addresses.

Finally, Fish's terminology is arguably misleading: hypotaxis is not crucially characterized by the use of connectives, but by the use of explicit grammatical embedding, hierarchical structuring that "places" (= -taxis) some words or phrases "beneath" (= hypo-) others. (And Obama's inaugural involves significantly more hierarchical embedding than Bush's 2005 speech did.)

Let's look a little more closely into the issue of counting sentential connectives. George W. Bush's 2005 inaugural contains 100 sentences, by my count, and 10 of them begin with "connectives that point backward or forward". In comparison, Barack Obama's 2009 inaugural contains 108 sentences, 14 of which involve sentence-initial connectives. I interpret this to mean that both addresses are written in a style that avoids explicit indications of discourse structure — but Obama's address doesn't have this property to any unusual extent. (To get agreement about the details of such an analysis, we'd have to spend some time deciding what counts as a "connective" of the appropriate sort, and discussing the interpretation of individual cases — but I'm very skeptical that any revision of this analysis is going to show a big difference between Obama's text and other recent inaugural addresses. For example, feel free to divide the addresses into tensed clauses and count the types of connections among those — I don't think you'll find support for Fish there either.)

As for the alleged lack of any "logical progression" between the "meditative beads" of Obama's 2009 inaugural address, I'm at a loss to understand what Prof. Fish has in mind. Here's the first sentence of the third paragraph, which is the real start of the speech.

Forty-four Americans have now taken the presidential oath.

Now the second sentence:

The words have been spoken during rising tides of prosperity and the still waters of peace.

There's no explicit connective or adverb indicating the discourse relation — no and so or therefore or nevertheless. But "the words" refers back to "the presidential oath"; and the second sentence as a whole expands on the idea of the historical progression of presidencies that was introduced in the first sentence, evoking those that began in peace and prosperity.

The third sentence starts with the contrastive discourse-structural adverb "yet", and continues with a time-adverbial "every so often":

Yet, every so often, the oath is taken amidst gathering clouds and raging storms.

And the subject, "the oath", again carries the thread of the discourse forward.

At these moments, America has carried on not simply because of the skill or vision of those in high office, but because we, the people, have remained faithful to the ideals of our forebears and true to our founding documents.

This fourth sentence starts with an adverbial "at these moments" that refers back to the just-discussed times of clouds and storms.

Please feel free to go on through the speech in the same vein, and see if you can tell me whether it has any objective properties that Prof. Fish could have been responding to as "meditative beads" of parataxis. But before you tell me about your findings, please also figure out how your analysis distinguishes Obama's 2009 inaugural from (for example) Bush's 2005 address, which starts like this:

At this second gathering, our duties are defined not by the words I use but by the history we have seen together.
For a half a century, America defended our own freedom by standing watch on distant borders.
After the shipwreck of communism came years of relative quiet, years of repose, years of sabbatical, and then there came a day of fire.

The rhetorical approach is similar: the challenges of the moment are placed in historical context. The first sentence of Bush's paragraph tells us that "the history we have seen together" is crucial. The second and third sentences lay out three periods of that history: the cold war, the lull after the fall of the Soviet Union, and the aftermath of 9/11. However, this structure seems to me to be less explicitly signaled (by connectives, adverbs and anaphora) than the structure of Obama's paragraph was.

In both addresses, the phrases and sentences have a clear structure as a coherent discourse, easily accessible to any sentient reader. In neither case are we presented with the sequence of disconnected "meditative beads" that Prof. Fish claims to find in Obama's speech. In both cases, much of the discourse structure is implicit, requiring us to understand the content in order to grasp the form — but Obama's speech seems not at all unusual in this respect.

What about the interpretation of parataxis as flat structure, contrasted with hierarchical embedding? Can we rescue Fish's response by looking at the speech from that perspective? I discussed the stylistic history of "Inaugural embedding" a few years ago (9/9/2005). In that post, I compared the "number of words at each level of embedding" in three inaugurals — George Washington's, Abraham Lincoln's second, and George W. Bush's second. (See the post for how I operationalized "level of embedding", and other details.)

Expanding the table from that post to include Obama's speech, we get:

	0	1	2	3	4	Mean Sentence Length
GW1	629 (44%)	554 (39%)	206 (14%)	36 (3%)	5 (<1%)	60
AL2	440 (63%)	222 (32%)	38 (5%)	0	0	26
GWB2	1842 (88%)	244 (12%)	4 (<1%)	0	0	22
BO1	1493 (62%)	799 (33%)	101 (4%)	0	0	22

Thus by this measure, Obama 2009 was again less paratactic than Bush 2005 was. (In fact, his degree-of-embedding proportions are quite similar to those of Lincoln 1865.)

My conclusion: under deadline pressure, Prof. Fish reached in his bag of analytical concepts and pulled one out, more or less at random, that expresses something about his response, but has little or no specific connection to Obama's text. In his interpretive community, the term parataxis has some broader resonances that may help explain why it came to his hand. Thus Charles Bernstein ("What's Art Got to Do with It? The Status of the Subject of the Humanities in the Age of Cultural Studies", American Literary History 5(4): 597-615, 1993):

I agree with Fish and others that the new interpretive approach changes the objects of study. […]

The poetic — the aesthetic — the philosophic — the rhetorical: these intertwined figures dissolve into the art of everyday life, the multiple and particular decisions and revisions, recognitions and intuitions, that make up — constitute — our experiences of and in the world. […]

Such hypertextuality offers not a theory of frames — a supervening or hypotactic ordering principle — but an art of transition through and among frames. Call it the art of parataxis, where the elements set side by side are critical methods rather than images or ideas: an art of practice, which provides not answers but paths of reading and provisional connections among these paths.

I sympathize, but it seems to me that the analysis of Barack Obama's text as particularly paratactic is still, in the philosophical sense of the term, bullshit.

[Some other relevant LL posts: Discourse: branch or tangle?", 11/6/2003; "Homo hemingwayensis", 1/9/2005; "The evolution of disornamentation", 2/21/2005; "Parataxis in Pirahã", 5/19/2006; "Sex & language stereotypes through the ages", 9/10/2006; "Baseball conditionals", 5/23/2007; "Roeper on recursion", 6/24/2007; "A principle that no one can convince me that doesn't exist", 8/18/2007; "Ontological promiscuity vs. recursion", 2/10/2008.]

January 24, 2009 @ 11:46 am · Filed by Mark Liberman under Language and politics

Permalink