Monday, September 24, 2012

A Better Text, Really? Shades of Mike Hancher

Harman’s recent suggestion that somewhere in the metaphysical ether there’s a text better than the one(s) written by the author reminded me of a piece one Michael Hancher published back in the ancient days when I was an undergraduate and then a Master’s student at Johns Hopkins and Hancher was a newly hired assistant professor of English. But I don’t want to go directly to Hancher’s text, in which he suggests that the art of criticism resides in (the possibility of) finding a meaning for a text better than the author’s meaning, which is the province of the science of criticism. I want to sidle up to it.
Note that the next section gets a bit sticky once I invoke the distinction between signifiers and signifieds. Alas, that cannot be helped.
What’s Special about Literary Texts as Objects?

Let’s reprise a passage from Harman’s reply to Green:
...I think the literary text is something deeper than its current holistic configuration in the form of how the author chose to publish it, or the best available scholarly version of a text available at any given moment, or whatever is usually taken to be the real text. ...

In other words, I think I’m giving an even stronger critique of authorial intention than is usually the case. Not only do authors fail to master the infinite dissemination of their texts, they probably don’t even put the text in the right shape in the first place. Most of them should have written better texts. Just as social surroundings fail to exhaust a literary work, the exact written form of a literary work fails to exhaust the deeper spirit of that work. My article proposed the rudiments of some methods to get at that deeper spirit.
Let me again ask: What is this unwritten text of which Harman speaks, the one that authors fail to shape properly? I have already suggested, in Is Harman a Platonist?, that Harman is taking advantage of the ambiguity and equivocation inherent in the notion of the text as a term of art in literary study. While the text certainly includes the written symbols of the text, assuming we’re dealing with a written text, there’s some ambiguity even here. The exact written marks are likely to vary between manuscripts, if any still exist, and published editions, and the variations reflect mistakes and intentions by several parties: authors, editors, and printers. But the notion of the text extends from those visible, albeit often disputed, marks to some vague nebulous thing of which those marks are said to be an expression. It’s that nebulous thing, more often than not, which is the object of critical investigation. That’s what we mean by “the text.”

What I find so odd is Harman’s assertion that he’s “giving an even stronger critique of authorial intention than is usually the case” and that he’s giving it in favor of this better but vague and nebulous thing. How does he know this thing exists in any way shape or form? The fact that he can designate it by a suitable noun phrase – “the deeper spirit of that work” – doesn’t mean that that thing exists as anything other than a figment of Harman’s philosophical imagination.

I’m curious as to whether there are better hammers or rocks of this sort. Of course, given some hammer and a task for it to perform, there may well be a better hammer for that particular task. And, depending on circumstances, one might even be able to obtain that better hammer. But that’s not really what Harman’s getting at when he talks about the better text. At least I don’t think it is.

Harman talks of real objects as always withdrawing. The hammer we have at hand, however suited it may be to the current task, is not the real hammer. Rather, it is, in Harman’s terminology, a sensual hammer. And if, in a given case, we set aside the hammer most readily at hand and get a different hammer, well then that different hammer is but a sensual hammer as well.

Texts are more complicated than hammers, even without the metaphysics.

Linguists, and philosophers too, differentiate between the signifier, which is the physical sign, and the signified, the meaning of the sign, the physical basis of which is as yet somewhat obscure. Now, it is possible that when Harman invokes the deeper spirit of the work he means that the invisible structure of signifieds is that deeper spirit. This would imply that that invisible structure could be expressed by various different strings of signifiers. Harman would thus be treating those invisible signifieds as the real object, the one that’s ever withdrawing.

Consider Shakespeare’s two versions of King Lear, the 1608 Quarto version called The History of King Lear and 1623 Folio version called The Tragedy of King Lear. It would be an abuse of the linguistic concepts, signifier and signified, to think of these texts as two strings of signifiers corresponding to one string of “deeper spirit” signifieds. No. We’ve got two strings of signifiers and they express two structures of signifieds. Whatever the deeper spirit of King Lear is, it can’t be the structure of signifieds invoked by either of those two texts, nor the one invoked by Nahum Tate’s 1681 redaction.

I assume that by “deeper spirit” Harman means the real text, the inaccessible one, the perpetually withdrawing object. In the case of King Lear, the fact it has given rise to at least three closely related strings of signifiers implies that it cannot be identified with any singles structure of invisible and, alas, all but-inaccessible signifieds. In each case, then, we have one text we can see, the string of signifiers, and two that we cannot, a corresponding string of signifieds and that deeper spirit that Harman believes animates these three versions, and others as well.

What I’m suggesting is that Harman has inadvertently conflated his notion of a real object (considered as something inaccessible and withdrawn) with the somewhat different notion of a structure of signifieds underlying a given text. When dealing with hammers, and rocks, lakes, quarks, galaxies, aardvarks and mustard seeds, we don’t have to worry about a distinction such as that between signifier/signified. And so, when dealing with such objects there’s no possibility of conflating the real object with a structure of signifieds. Literary texts are not so simple. It’s not at all clear to me that OOO has the conceptual equipment to distinguish between invisible signifieds of visible signs and real objects in endless withdrawal from sensual objects.

Where’s this “Better” Text?

Let’s continue on, but in a more concrete mode. In this more concrete mode you may assume, if you wish, that Harman is able to overcome the argument I’ve presented in the previous section and that the notion of a deeper spirit is alive and well-distinguished from that of a structure of signifieds. Still, for this notion to be of any practical value we need some evidence for it beyond Harman’s assertion.

The best sort of evidence would be the existence of a text that’s better than the author’s original. Harman offered some suggestions in this direction his original article, namely:
Namely, the critic might try to show how each text resists internal holism by attempting various modifications of these texts and seeing what happens. Instead of just writing about Moby-Dick, why not try shortening it to various degrees in order to discover the point at which it ceases to sound like Moby-Dick? Why not imagine it lengthened even further, or told by a third-person narrator rather than by Ishmael, or involving a cruise in the opposite direction around the globe? Why not consider a scenario under which Pride and Prejudice were set in upscale Parisian neighborhoods rather than rural England—could such a text plausibly still be Pride and Prejudice? Why not imagine that a letter by Shelley was actually written by Nietzsche, and consider the resulting consequences and lack of consequences?
As I remarked in my original critique, these suggestions collapse into the actual processes of literary culture. Abridged versions of classic texts abound and stories are routinely transposed to different settings. Even if Pride and Prejudice hasn’t been reconstructed in upscale Paris—and maybe it has—such things have been done. Akira Kurosawa, for example, has re-imagined Macbeth as Throne of Blood and King Lear as Ran.

But Harman’s original article said nothing about authors writing better texts than they have in fact done. If Harman is serious about “better” than he is invoking the very problematic business of evaluation. Who’s going to be the judge, and what criteria will they use?

Is Ran better than any King Lear text, or just different? Is The Tragedy of King Lear, the 1623 version of Shakespeare’s text, better than The History of King Lear? Alfred Harbage, editor of The Pelican Shakespeare (which I used in graduate school) thinks so, calling it “greatly improved” though “cut.” But the version Harbage himself prepared is based on both texts, as are almost all modern versions. Presumably he thinks a judicious combined text is better than either of the originals—if it’s appropriate to call them that.

Nor am I in any position to question him as I’ve not read either of the originals. Harbage also notes that he uses act and scene divisions from the Folio text and that the Quarto text has no such divisions at all. Moreover notes that “none of the plays printed in Shakespeare’s lifetime [the Folios were printed after his death] were divided into acts and scenes, and the inference is that author’s own manuscripts were not so divided.”

Now we have a problem. In 1972 Mark Rose published Shakespearean Design, in which he argued that the act and scene divisions obscure the structure of the plays. He also pointed out that the Elizabethan stage had very little in the way of scenery, so that it was unnecessary to stop the action between scenes or acts in order to change the scenery. Thus the performances moved more quickly in Shakespeare’s time than in hours. The thrust of his argument is thus that these modern improvements have degraded the plays.
For what it’s worth, Norman Holland has observed (in an email to the PsyArts listserve) that he’s found sound recordings of Shakespeare plays to be the most persuasive recorded versions. Such recordings do not, of course, take time to change sets between scenes.
As far as I can tell Rose’s argument has had little or no impact, though I’m partial to it myself. My point though is simply that the addition of this business of value, of a better text more true to a deeper spirit, makes a complicated and subtle business—what is the text?—even worse. If Harman is serious about that idea, then we need to have at least some of those better texts. Otherwise it’s an empty idea. But judging these counter-factual texts to be better, that’s going to be a mess.

Let me conclude by quoting a passage from the dedicatory epistle with which Nahum Tate prefaced his 1681 redaction of King Lear:
The Images and Language are so odd and surprizing, and yet so agreeable and proper, that whilst we grant that none but Shakespear cou'd have form'd such Conceptions, yet we are satisfied that they were the only Things in the World that ought to be said on those Occasions. I found the whole to answer your Account of it, a Heap of Jewels, unstrung and unpolisht; yet so dazling in their Disorder, that I soon perceiv'd I had seiz'd a Treasure. 'Twas my good Fortune to light on one Expedient to rectifie what was wanting in the Regularity and Probability of the Tale, which was to run through the whole A Love betwixt Edgar and Cordelia, that never chang'd word with each other in the Original.
Tate clearly intended to improve upon Shakespeare. History has not smiled on his intentions and his text has been forgotten by all but a few academic specialists. Perhaps history has judged wrong, but who’s going to make that call?

The Best Meaning May Not Be the Author’s

At this point I have produced one argument asserting that Harman’s invocation of a “deeper spirit” is mistaken, perhaps because he conflated the real text with the structure of signifieds underlying the string of signifiers that is the physical text. I’ve produced another argument which asserts first that for Harman’s argument to be meaningful, we have to see better texts. I then point out that making such a value judgment will be difficult and I offered the mess of texts pretending to be the real King Lear as evidence of what actually happens in literary history. And these multiple versions have happened without the promptings of object-oriented philosophers. King Lear may be an extreme case, but it’s not unique.

I’m now going to argue that Harman’s plea for a better text has been oddly anticipated by Michael Hancher in 1970. Hancher wasn’t writing about the texts themselves (whatever those texts are as objects), but only about interpretations of texts. Given the previous arguments, however, it’s not at all clear to me that that’s much practical difference between Hancher’s argument and Harman’s.

What interests me is a particular conceptual move that Hancher makes in the course of his argument.

The essay is called The Science of Interpretation and the Art of Interpretation and it appeared in Modern Language Notes (Vol. 65, No. 6, 1970, pp. 791-802). Hancher compares Cleanth Brooks with F. W. Bateson on Wordsworth’s “A Slumber Did My Spirit Seal.” Of that comparison he asserts (p. 796) that
I will persist in finding that Brooks’ reading makes a wholly more satisfactory work out of Wordsworth’s poem than Bateson’s does. Bateson’s reading may be the more valid reading, but Brooks’ poem is the better poem—it is as simple at that.
Of course, Hancher provided reasons for Brooks’ reading over Bateson’s. But I’m not interested in those reasons; I’m willing to accept them at face value. I’m interested in how Hancher then goes on to argue that it is admissible to prefer a reading other than one that could reasonably be attributed to the author.

As a matter of definition Hancher asserts that, in arguing on the basis of authorial intention, Bateson was practicing the science of interpretation while Brooks was practicing the art. Hancher wants to defend the art. And he makes the distinction because of the kinds of objects he’s dealing with, natural objects vs. man-made (i.e. literary texts).

Given the distinction, however, Hancher then quotes a remark that Oscar Wilde had made about Wordsworth, that he “found in stones the sermons he had already hidden there.” Hancher then remarks:
Wilde’s joke at Wordsworth’s expense points to one fact about interpretations of which Wordsworth himself was acutely aware: if natural objects are merely neutrally existent, uncreated things, and not divine artifacts, whatever meaning we may find in them we will have in fact projected onto them ourselves. If nature is not a divinely authorized book, no efforts to interpret can be scientific. Only an art of interpretation can generate and justify meanings that are not meant.
Scientifically, as naturally occurring objects, stones have no meaning. But one can, if one so chooses, view stones with other than a scientist’s eye. One can artfully attribute meaning to them.

And now we come to a curious twist:
Another way to put the matter would be to say that the art of interpretation is appropriate chiefly to unintended, natural objects. But it will also embrace intended, created objects if they are considered as if they were unintended, natural objects.
As far as I can tell, this particular art-science distinction has no meaning in object-oriented ontology. What’s important is that OOO treats all objects alike, whether natural or man-made. And that’s what Hancher proposes to do, to treat (certain) man-made objects, poems, as though they were natural objects.

What’s interesting is that Hancher ends up against the nature/culture divide, as does OOO, and chooses to dissolve it, as does OOO, and does so because he wants to improve upon the author’s text, as does Harman.


Hancher has various meanings chasing the text, including some better than the author’s meaning. Harman has various texts chasing the deeper spirit, including some not written by the author.

The parallel is interesting, but curious.

Hancher regards the interpretations as some form of knowledge. It’s not clear where Harman purports to find knowledge in his counter-factual manipulation, or even if he expects to find knowledge at all.

And this is intellectual progress.



  1. well if the author wrote that in-tune with themselves with genuinness then that's the best we can get

    all the other improvements will necessarily be brainy (is literature to cherished as a technology object?) and thus they better wrote their own

    that's what you get when you mix technology with philosophy you get all these absurdities or impasses

    Harman is a transcendentalist (as you indicate with a slight (or more) tinge of negative dissociation

    1. Actually, I think Harman is sometime a transcendentalist and sometime an immanentist, depending on circumstances. It may all seem immanent to him, but he's not the only judge of his prose. It's that old gestalt thing: duck or rabbit? beautiful lady or old crone?