In the first post in this series I addressed the nature of minimal reading as Attridge and Staten presented it in the 2008 web version of their first chapter, on William Blake’s The Sick Rose. Now that I’ve worked my way through their treatments of five poems – The Sick Rose, Lenox Avenue: Midnight, I started Early, At a Solemn Music, and Futility – I return to that general discussion, but in a somewhat quizzical and critical vein. I want to address these remarks to statements from their introduction.
Dialog and Intersubjectivity
Here’s their second paragraph (p. 1):
As we gradually saw, there is more to the dialogue form of our expositions that we initially realized – something new, perhaps, in poetry criticism; we could call it dialogical poetics. We are accustomed to seeing all extended commentary on poetry as the vision of some individual consciousness – the more individual and “original,” the better. There is almost always a significant degree of arbitrariness in the pronouncements of such solitary readers: associative leaps, inferences drawn, symbolic meanings perceived, that might be more or less plausible, but the justification of which we are left to figure out for ourselves if we can. In this book, by contrast, the two authors have had to justify their readings to each other, step by step, and we have left visible the process by which we worked through our perceptions to reach at least partial agreement.
They seem to think there is some necessary, or at least very strong, connection between the dialog form and minimal interpretation. This strikes me as peculiar because I have, for years, been working “close” to the text as a solo practitioner. I think of my work as descriptive rather than interpretive. I can do that without having to enter into dialog with another critic and I do it under the assumption that what I’m describing is really there and thus available to all readers, whether professional or not, and regardless of what they consciously think.
I understand the value of collaboration – I worked closely with David G. Hays for two decades and we cosigned a small handful of papers – so I certainly have no objection to critical dialog. Moreover, I am in some sense in conversation with Attridge and Staten in these posts even if I am not writing directly to them in email. I’ve certainly benefitted from their comments on the poems to which I addressed myself. But they seem to have more in mind than mere, if you will, collaboration.
Because dialogical poetics forces Attridge and Staten to explain themselves to one another they claim for it an epistemological value that is lacking in (now) standard critical practice which, in their characterization, values critical originality and features somewhat arbitrary assertions that are poorly justified. While this makes sense, there’s something peculiar going on. Why is mutual intelligibility, intersubjectivity, such a big deal for them? It seems to me that they’re taking the long way around to a statement about the collective nature of poetry and not quite getting there.
So, for the sake of argument, let us assume that canonical texts function as devices – “cultural beings” as I have called them in other contexts – which allow groups of people to arrive at shared norms and values. If that is so, then it is rather peculiar to have a small elite group take upon themselves the task of tending these texts, these cultural beings, by competing at explicating those texts in arcane and creative ways. It is even more peculiar that these competitions are often governed by critical systems asserting that these obscure and arbitrary explications reveal how these texts hold groups of people captive to oppressive value. How can an elite intellectual sport contribute to the collective value of canonical texts? But also, how can texts whose meanings are indeterminate hold anyone captive to anything?
The purpose of the dialogic pursuit of minimal interpretation, then, is to “reverse” the “energy” in this system so as to produce a critical practice that is more consistent with the collective functions and responsibility of the literary system. I’m just not sure we can get there from here.
Grammar and Poetics
Now we have a long paragraph in which Attridge and Staten counterpoint their enterprise with that of linguistic analysis of sentence-level grammar. As I have quite a bit I want to say about this paragraph I think the best way to proceed is first to present the paragraph without comment and then to repeat it in full, this time inserting commentary along the way. Here’s the full paragraph (p. 3):
We are interested in poems as linguistic artifacts, as made things that manifest the techniques of making by means of which poets produce them at specific times and places. No one would try to interpret a poem without knowing the grammar of the sentences in it; and yet it is common practice to interpret poems with only the vaguest sense of the poetic techne that organizes it formally, at a different level, but just as necessarily, as does grammar. The formal organization of a poem is so complex and multi-leveled that no two critics are likely to analyze it in exactly the same way, but then, there are many different ways of analyzing and representing the grammar of a language. This is what bothered Noam Chomsky about traditional grammars and led him on his quest to develop a “perfectly explicit” grammar, a quest that remains unfulfilled; yet no one thinks that grammatical analysis is a merely arbitrary product of interpretive methods. If the lack of absoluteness is the essential characteristic of interpretation, then the kind of structural analysis we do is also interpretation; but its aim is to be as minimally interpretive as possible. Thus its protocols dictate a closer adherence to the text, in all its details and as a whole, than do other approaches.
This paragraph seems OK as long as you don’t think about it very deeply. Once you do, however, it begins to unravel. I suppose that, in pointing this out, I might be accused of reading them uncharitably. If so, well, then I’m uncharitable. With that in mind, let’s give it a go:
We are interested in poems as linguistic artifacts, as made things that manifest the techniques of making by means of which poets produce them at specific times and places.
Yes, of course, linguistic artifacts and how they are made. On the one hand there are techniques of which we are consciously and explicitly aware, which may even have names, and which can even be deliberately taught. But there are also things that we do consistently, but of which we have little or no deliberate awareness. Language partakes of both.
No one would try to interpret a poem without knowing the grammar of the sentences in it; and yet it is common practice to interpret poems with only the vaguest sense of the poetic techne that organizes it formally, at a different level, but just as necessarily, as does grammar.
What do they mean by “knowing the grammar of the sentences”? What kind of grammar are they talking about? We have the sort of informal grammar taught in primary and secondary schools but we also have the sort of grammar created by professional linguists in an effort to understand how language works. While these two types of grammar share vocabulary – e.g. such terms as subject and object, verb and noun, and so forth – they are quite different constructs. These days the professional grammars often make use of formal expressions (derived from mathematics) and formal diagrams. As few literary critics are conversant in the sorts of grammars professional linguists have developed over the past half-century, I assume that they are referring to the informal grammars taught in primary and secondary school.
I agree with them that it is common to interpret poems without knowledge of poetic techne. However, it is not as though poetic techne has been codified even as well as informal sentence-level grammar. There’s prosody, which apparently isn’t much taught any more, and classical rhetoric, ditto, and that’s pretty much it. For the most part when Attridge and Staten discuss poetic techne in the book, they’re making it up as they go along, rather than calling on well-developed concepts that have simply been forgotten. That is fine, of course, but the parallel with sentence grammar is not very good. Informal sentence grammar is well developed, but professional linguists are in deep disagreement about more rigorous descriptions.
The formal organization of a poem is so complex and multi-leveled that no two critics are likely to analyze it in exactly the same way, but then, there are many different ways of analyzing and representing the grammar of a language.
The parallel they’re offering seems weak to me. Characterizing the organization of a (single) poem strikes me as rather different from characterizing the grammar of a language which, in the style championed by Chomsky, involves accounting for an infinite number of sentences and doing so with a quasi-mathematical formalism. And the differences in “analyzing and representing the grammar of a language” exist between different schools of linguistics more than at the level of individual linguists.
This is what bothered Noam Chomsky about traditional grammars and led him on his quest to develop a “perfectly explicit” grammar, a quest that remains unfulfilled; yet no one thinks that grammatical analysis is a merely arbitrary product of interpretive methods.
That strikes me as a rather odd way of characterizing what Chomsky was (and is) up to. It is so odd, in fact, that I don’t quite know how to argue against it.
Explicit, sure, but that was hardly new with Chomsky. He was driven by computational considerations, such as the need to account for an infinite set of sentences over a finite collection of words and rules. And, as I’ve already indicated, he expressed that grammar in formal expressions and diagrams. He – and other linguists – have created specialized technical means for talking about language, aka linguistic artifacts, while Attridge and Staten propose nothing of the kind in their critical work in this book. They reject such (quasi) technical language as literary critics have developed – for those systems are incompatible with the basic aim of weak interpretation – though they do make some use of (traditional) prosody.
In any event, the first clause of the Chomsky sentence exists to set-up the second one, which is motivated by a problematic that has been driving (academic) literary criticism since the 1960s, the notion (and fear) that literary critics see in texts only what a given interpretative system allows them to see – and there’s no way out. But it’s not clear to me that “arbitrary” is a good word here. After all, each interpretive system has a rationale; they are not capricious intellectual contraptions, though some may be better crafted than others.
As for grammatical analysis, yes, different schools of linguistic thinking will impose different methods of grammatical analysis; that’s the point of having different schools. But “arbitrary” isn’t a useful way of characterizing the differences between schools. Just as each literary critical school has its reasons, so each linguistic school has reasons for its methods. Those reasons weren’t simply drawn from a hat. They reflect differences in fundamental conceptual outlook.
If the lack of absoluteness is the essential characteristic of interpretation,
But is it THE essential characteristic? That assertion strikes as quite odd.
then the kind of structural analysis we do is also interpretation;
Are they telling us that their first consideration in choosing what to assert about a text is that it not be absolute? Again, that strikes me as being odd.
but its aim is to be as minimally interpretive as possible. Thus its protocols dictate a closer adherence to the text, in all its details and as a whole, than do other approaches.
I’m all in favor of attending to the text. I just don’t see why they’ve employed an analogy to sentence-level grammar in linguistics. What they actually do in their analytic work is useful and insightful. Whatever reservations I may have about their work – I’d like to see more attention to describing formal features including global features that resist being explicated in the language of meaning (e.g. ring-composition in the Langston Hughes, overall rhyme patterns in the Dickinson and the Milton) – I would certainly like to see more work like this.
Let me quote one more passage that raises these kinds of issues (p. 4):
We can think of minimal interpretation as a kind of “reverse engineering.” When a mechanical or electronic device is reverse engineered, what we understand we understand is how it is put together in such a way as to function as it does. There are many other contexts other than that of its functional organization in which a device can be understood; but the context posited by reverse engineering is on a plane that is sui generis.
It sticks in my mind that notion of “reverse engineering” entered cognitive psychology in the late 1950s through Donald Broadbent, though I have no citation to offer. More recently Steven Pinker used it in his 1997 How the Mind Works. But what Attridge and Staten are doing is little or nothing like the models and mechanisms posited in linguistics and the cognitive sciences. The cognitive sciences are not interpretive except in the large, vague, and useless sense that everything is someone interpretive. That is, once again Attridge and Staten are invoking the cognitive sciences in a way that bears little relation to what they’re actually doing in this book.
Whatever it is that they’re doing, whatever this dialogic weak interpretation is, it is not reverse engineering the “generative” techne of poetry. What can that possibly mean? Whatever it is that they HAVE got, they don’t have the sense of mechanism that underlies much of the best work in linguistics and the cognitive sciences. What I cannot tell is whether they’re indifferent to the sense of mechanism or whether they’re actively resisting it.