Friday, March 18, 2016

Notes on Attridge and Staten: The Craft of Poetry

You can download the complete working paper here:

Abstract, contents, and introduction below.


Abstract: In The Craft of Poetry Attridge and Staten propose a method of minimal interpretation that they illustrate through dialog with one another. In minimal reading one dispenses with theory-driven methods and seeks to come as close to a literal interpretation of the poem as possible. Moreover by dialoguing with one another Attridge and Staten force themselves to justify their interpretations in explicit terms. I offer general methodological commentary and comment on their treatment of five poems: William Blake, The Sick Rose; Langston Hughes, Lennox Avenue: Midnight; Emily Dickinson, I started Early; John Milton, To a Solemn Music; and Wilfred Owen, Futility.


0. Introduction: The Limits of “Reading” 2
1. What is Minimal Reading? 9
2. Figuration: The Sick Rose 14
3. Formal Features of The Sick Rose 25
4. Langston Hughes Crafts a Ring 28
5. Dickinson started early, turned a figure with the sea 31
6. Some Notes on Milton’s Solemn Music 36
7. Wilfred Owen’s Futility and the issue of historical context 42
8. What (do they think) they are up to? 46
9. Reading, Meaning, Techne 51

Introduction: The Limits of “Reading”

This working paper consists of nine posts from my blog, New Savanna, which I’ve recently written about The Craft of Poetry: Dialogues on Minimal Interpretation (2015), by Derek Attridge and Henry Staten. Three of them, the first one and the last two, are about method and theory – “theory” in the general sense, not in the peculiar sense it assumed in literary studies in recent decades. The other six posts engage Attridge and Staten on five of the poems they examine in the book.

Read it, but…

Just what I think about this book, that is tricky. On the one hand I recommend it to anyone with a serious interest in the analysis of poetry. It exhibits useful analytic skills and delivers fascinating accounts of an interesting range of texts. And it is accessible to those with relatively little experience (such as undergraduates) while also giving the most experienced professionals things to think about. What’s not to like?

And yet, as I ask in my penultimate post: What do they think they’re up to? Yes, they are offering remedial instruction, perhaps even deeply remedial, but they also write as though their method of dialogic poetics, as they call it, offers the profession a new departure, and of that I’m skeptical. I can see some special journal issues being devoted to dialogic poetics. In fact I think it would be a fine idea – can you imagine, say, an issue of ELH in which every article address a text or three, no more, and multiplied authored by critics in conversation with one another? For that is how each chapter of The Craft of Poetry is written; Attridge and Staten address remarks, questions, and responses to one another. I can also imagine some conferences in the same vein, with presentations by two or more critics and perhaps even some “open” sessions where a text is proposed and those present discuss it in real time – I’m thinking of those marvelous sessions that Haj Ross had at his languaging conferences at the University of North Texas back in the 1990s. But I can’t really see much more than that. I can’t see it becoming a way of professional life.

The problem is that Attridge and Staten do not question what is in effect the unstated but foundational assumption of academic literary criticism, the primacy of discursive thinking [1]. This assumption is maintained through the trope of “reading” – the term used to assert continuity between poetry (or any literary text) and the analysis and explication of poetry, as though they are essentially the same mental act. The pull of this trope is so strong that Franco Moretti even adopted it for forms of computational criticism that are obviously discontinuous with reading, as the term is ordinarily understood, by virtue of the fact a critical phase of the analytic procedure is carried out by a digital computer. Distant reading, Moretti’s term, simply is not a form of reading at all [2].

Any form of explication, analysis, or interpretation, by whatever method, even the “minimal” (my scare quotes, not theirs) method Attridge and Staten employ is a secondary, a derivative, activity that is not continuous with poetry itself. Poetry leaves one kind of footprint in the mind and the world, while its analysis does something else. The relationship between these two kinds of work is not at all obvious, hence Archibald’s MacLeish’s assertion that “Poems should not mean/but be.” The trope of reading undermines that assertion through the trope of distance. When we undertake a close reading (a term, by the way, that Attridge and Staten explicitly reject) we are so close to the text and we might as well be reading it. And when we use a computer to crunch over 1000 texts, well, we’re no longer very close to any of them. And so we assert distance, and that enables us to pretend that it is still a kind of reading and so essentially continuous with those texts.

We can see this problematic played out in the title essay of Geoffrey Hartman’s 1975 collection, The Fate of Reading. In that essay Hartman is grappling with the implications of semiotics and linguistics for literary criticism. Complaining that contemporary theorists—mostly French or those under French influence—have come to privilege analytic writing over reading, Hartman asks (p. 272): “To what can we turn now to restore reading, or that conscious and scrupulous form of it we call literary criticism?” That is, how can we make our (necessarily written) critical practice continuous with the experience of reading texts? He then observes: “modern ‘rithmatics’ – semiotics, linguistics, and technical structuralism – are not the solution. They widen, if anything, the rift between reading and writing.” I believe that Hartman is correct but I do not share his nostalgic delusion one can get “closer” to the text through the proper method. Moreover I regard the rejection of the world of linguistics and technical structuralism as a mistake.

Moreover I note that that world is one where discursive thinking is augmented by some other mode – or modes – and that augmentation is essential. It may involve diagrams – think of all the diagrams Lévi-Strauss produced – or mathematical and logical formalism, as in Chomskyian linguistics, or empirical investigation. Whatever it is, it is essential to the enterprise. The investigation must go beyond the bounds of discursive thinking. I suspect, though obviously I cannot prove, that it is the importance of these other modes of thought that bothered Hartman. They intruded on his illusion of closeness.

In search of the “generative system of poetry”

With that in mind let’s take a look at a passage from the introduction to The Craft of Poetry (p. 2):
Our aim in publishing our dialogues on poetry was to make a case for certain skills of poetry reading – including prosody – that we believe constitute basic poem literacy, and which over the past four decades have been shoved aside in many literature departments [...] we wanted to demonstrate how poems can be read based on the assumption that it is not, in the first instance, theories, “interpretive communities,” readerly competence, or historical forces, but poets, and behind poets the techne (art or craft; art considered as craft), or generative system of poetry, that “produces” poems.
That is all well and good, but just what do they mean by “generative system of poetry”?

That is not at all clear. Later in the introduction they’ll invoke Chomsky’s work on grammar as the sort of thing they’re after, except of course that they are interested in whole poems, not sentences. The trouble is that what they do with poems is nothing like what Chomsky was doing for sentences. Similarly, they invoke the notion of “reverse engineering,” which Steven Pinker has used in explicating cognitive science. But again, what they do is nothing like that – I develop this criticism more fully in my eighth and ninth posts, starting on pages 45 and 50 respectively. They don’t even attempt to summarize what they’ve learned through analyzing these poems. Their nascent grammar of poetry is thus empty of propositions about poetry.

In a sense, though, this is a minor matter. So they’ve mischaracterized their enterprise. As long as what they actually do is interesting, who cares?

Well, I care, for I’ve spent a great deal of time thinking about the “generative system” that “produces” poetry, and in terms that are commensurable with the world of the cognitive sciences. That’s a deep problem, and it seems much deeper to me now than it did when I first broached it years ago, hoping to find the system that showed the “hidden grammar” (my scare quotes) linking Coleridge’s conversation poems to “Kubla Khan.” That project fell apart on me some time ago, but in the process I’ve arrived at some well-considered views about how to proceed.

You can find that story in a long blog post addressed to one of my graduate school professors, Literary Studies from a Martian Point of View: An Open Letter to Charlie Altieri [3]. The upshot is that I’ve come to believe we have to start with a careful description of the formal characteristics of literary texts. While that is an aspect of A&S’s method it’s not a focal point for them as it is for me. Consider one of the poems they discuss, Langston Hughes’s Lenox Avenue: Midnight. It’s relatively short, only 14 lines, but lines 7 and 8 are structurally central, and not merely numerically central (see my discussion starting on page 27). When you look at elements before and after those lines you realize that they balance one another, like this, A B C X C’ B’ A’, where X is lines 7 and 8. They thus exhibit ring-composition, a phenomenon that is relatively well known, at least in certain circles, but not, so far as I can tell, mainstream literary criticism.

Mary Douglas, the great anthropologist, devoted her last book to the subject, Thinking in Circles: An Essay on Ring Composition (Yale UP 2007), reviewing work that had been done on classical and Biblical texts, but also arguing that the practice is by no means captive to the ancient world. Over the past several years I’ve been looking for ring-composition in more modern texts. The Langston Hughes is not the only one I’ve found. There’s Joseph Conrad’s Heart of Darkness, albeit in a reduced form, and the Japanese creature-feature, Gojira. President Obama’s Eulogy for Clementa Pinckney exhibits the form, as does the Walt Disney cartoon, The Sorcerer’s Apprentice. The two sections of Coleridge’s “Kubla Khan” are each center-symmetrical. There are a few other ring-form texts I know of and, I assume, countless others that have not been identified as such.

Of course, only some of these texts are lyric poems. But in a way that’s half my point. Here is a global formal feature that is exhibited by a wide variety of texts. What can we learn by comparing these different texts and seeing how they utilize center symmetry? Moreover it is a textual feature that doesn’t require the use of abstract theory. You just look at the text and describe what’s there.

Of course, it’s not quite as easy as is sounds. Texts are complicated objects. It is not at all obvious just what features to look for. Some of the features are prosodic, but others are syntactic or semantic. Yet, even where the features are semantic, it is not a matter of figuring out what they mean, of interpreting, but simply of comparing features that appear in different parts of the text and noting such patterns of usage.

So, that is one thing. If we’re going after the (underlying) system of literature, the place to start is form, not meaning. But there’s something else.

Who cares about meaning? An ethical imperative

And that is this: While interpretive commentary on sacred texts is quite ancient, interpretive commentary on literary texts is relatively new. It is only in the years since World War II that it has become the norm in the academy, and yet it seems like it has always already been with us. That illusion is made easier by using the verb to read to elide the distinction between reading, as the term is ordinarily understood, and interpretive commentary on texts. If they are coextensive then one can justify the commentary merely by pointing to the texts themselves: they are important, therefore the commentary is important as well.

I believe that interpretive commentary must justify itself on grounds other than the cultural importance of the texts. After all those texts did their work, whatever it is, for centuries, even millennia, without any apparent need for interpretive commentary. Why should it continue?

Moreover, as long as we pretend that interpretation is continuous with reading, we allow the interpretive text to borrow its authority from that of the (canonical) literary text, even in the case where the critic believes that meaning is indeterminate. We must explicitly separate the authority of the interpreter from that of the primary text. Only then can we adequately recognize the ethical nature of interpretation.

Interpretive criticism ethical discourse, albeit a (relatively new) mode of ethical discourse, where I am talking of ethics in the broadest possible sense, of ethos, a way of life. My touchstone here is Wayne Booth, The Company We Keep: An Ethics of Fiction (1988). Ethical criticism is about how we live, and how we are to live in the future. It is a way of conversing with one another about such matters. As such it is quite different an inquiry in the “generative system of poetry,” which is necessarily a technical inquiry requiring specialized intellectualized expertise.

Thus I read Attridge and Staten as undertaking ethical discourse, an understanding that, as far as I can tell, is consonant with their own understanding. Their commitment to dialog with one another, to arrive a mutual agreement about the poems under study, is fundamentally an ethical one. It is a commitment that recoups the ethical function that literary texts play in our lives, as shared sources of values, attitudes, and ideals. This sets them against the virtuoso critic for whom literary texts can become a vehicle for intellectual display.

That is fine. But it has little to do with Chomsky and it needs to be recognized as a relatively new kind of cultural discourse. It is not reading, but a kind of ethical discourse built on the explicit sharing of ideas arising from and through literary texts.

* * * * *

I note finally that I’ve not read the whole book. There are chapters on Shakespeare, T. S. Eliot, Denise Riley, and a concluding chapter on translation (Baudelaire, García Lorca, and Rilke) that I have not read.

The Posts

I wrote these posts over a period of three months or so. I’ve gathered them into this working paper in the order in which I originally posted them and have left them unchanged, except for the typo or other minor matter that I happened to catch. A number of them involved fairly extensive quotation from Attridge and Staten so as to give you a feel for how they go about their work.

1. What is Minimal Reading? – In which I address the nature of minimal reading and take my own hobbyhorse for a short run, suggesting that formal features are particularly worthy of attention.

2. Figuration: The Sick Rose – I elaborate on A&S’s discussion of binary oppositions and introduce the philosophical notion of category error into into their examination of (figurative language in) William Blake’s The Sick Rose. I elaborate on that through a short poem by John Hollander, Coiled Alizarin, and introduce the idea, associated with cognitive linguistics, that the Great Chain of Being is embedded in natural language semantics. Blake’s figuration systematically exploits these structures.

3. Formal Features of The Sick Rose – In which I point out that second person address is critical to the poem by rewriting it in the third person. I then argue that the poem exhibits a four-part trajectory: 1) the address (line 1), 2) the excursion (lines 2-4), 3) the return (lines 5-6), and 4) the exit (lines 7-8).

4. Langston Hughes Crafts a Ring – Langston Hughes’s Lenox Avenue: Midnight exhibits ring-composition in which lines 7 and 8 are structurally central and the lines before and after are symmetrical about them. This structural feature brings the poem within the orbit of a variety of texts – narratives, films, poems – that exhibit that global form.

5. Dickinson started early, turned a figure with the sea – Emily Dickinson’s I started Early has a rhyme scheme that divides its six stanzas into three movements of two stanzas each. It opens in an almost pointillist disposition of divers elements and concludes with that diversity yoked into a dance form.

6. Some Notes on Milton’s Solemn Music – Ordinary syntax divides this 28-line poem (To a Solemn Music) into two sentences, one including lines 1-24, and other consisting of lines 25-28. The rhyme scheme divides the poem in two, ll. 1-16 and then ll. 17-28 (consisting exclusively of rhymed pairs). That two-part division maps onto semantics in that the first part is about the celestial world with the second part is about the fallen world of earth. I suggest that the sometimes-ecstatic nature of musical experience may lie behind Milton’s employment of Christian doctrine in the poem, especially as that ecstatic experience involved suspension of one’s will.

7. Wilfred Owen’s Futility and the issue of historical context – Rather than undertake analytic work of my own, as I did for other poems, I concentrate on the question whether or not Owen’s poem can be considered a war poem. I invoke an old passage from Kenneth Burke’s “Literature as Equipment for Living” and suggest that poetic meaning is elastic and that “it is that elasticity that allows the texts to bring diverse individuals within communal range of one another.”

8. Attridge and Staten 8: What (do they think) they are up to? – In which I push back against statements in their introduction where Attridge and Staten invoke linguistics (Chomsky on grammar) and cognitive science (“reverse engineering”) as justifications for what they are doing. Whatever it is that they ARE doing, it has little in common with the cognitive sciences and thus the comparison is empty and misleading.

9. Reading, Meaning, Techne – While Attridge and Staten speak of the discovery of techne, poetic craft, as the end of minimal interpretation, I suggest that techne is in fact more expansive than, exceeds, and is different from meaning. There is thus an unacknowledged tension between meaning and techne that pervades their work.


[1] I explore this in my working paper, Prospects: The Limits of Discursive Thinking and the Future of Literary Criticism (2015), 72 pp. URL:

[2] I address this in a blog post from 2012, Distant Reading in Lévi-Strauss and Moretti, URL:

[3] New Savanna, blog post, December 17, 2015. URL:

[4] I list all my work on ring-composition in the post mentioned in note 3 and have some discussion as well. Here’s that list, with links to online sources:

Kubla Khan, Samuel Taylor Coleridge: lyric. URL:

The Cat and the Moon, William Butler Yeats: lyric. URL:

Metropolis, Osamu Tezuka: manga (Japanese graphic novel). URL:

Nutcracker Suite, episode in Disney’s Fantasia: film. URL:

Sorcerer’s Apprentice, episode in Disney’s Fantasia: film. URL:

The Pastoral Symphony, episode in Disney’s Fantasia: film. URL:

Fantasia, Walt Disney: the entire film. URL:

Heart of Darkness, Joseph Conrad: novella. URL:

Apocalypse Now, Francis Ford Coppola: film. The film is loosely based on Conrad’s novella, Heart of Darkness. URL:

Gojira, Ishirō Honda: Japanese film. Note: An Americanized version appeared under the title, Godzilla, King of the Monsters, with footage cut from the film, new footage added, and somewhat reordered. The net result is that the ring-composition of the original was destroyed. URL:

Pulp Fiction, Quentin Tarantino: film. URL:

To a Solitary Disciple, William Carlos Williams: lyric. URL:

Eulogy for Clementa Pinckney, Barack Obama: sermon. URL:

The Road Not Taken, Robert Frost: lyric. URL:

The Meaning of the Digital Humanities, Alan Liu: expository article. URL:

Of these 15 works only seven are verbal texts; one is a graphic novel; and the rest (8) are films. This reflects my general range of interests these days. Of the verbal texts, five are standard literary texts (four poems and a novella); one is a sermon; and one is a theoretical article on literary criticism. Of the films, three are episodes in a larger film (Fantasia) and one of them (Apocalypse Now) is based on the novella (Heart of Darkness).

No comments:

Post a Comment