Irony Continued: Undecidability and Polysemy, Truth and Error

The ongoing discussion of Paul de Man’s “The Concept of Irony” continues with a long comment by Joe, and my response here. Newcomers to the discussion should start at the beginning here and here. Because this post attempts to respond to Joe’s latest comment line-by-line, most of it will be of little interest to those who haven’t been following the discussion (is anyone following the discussion at this point other than Joe and myself?). One point of interest even out of context might be my attempt to lay out de Man’s concept of undecidability in parts 3-5, starting here.

That brings us up to this post, where I intend to respond, as I said, line-by-line to Joe’s latest comments. They are quoted in their entirety here, but check out the above link to the comment if you want to see them in their original, undivided form. For the sake of clarity, I’ve placed Joe’s quotes in italics and mine in boldface.

1. Some Initial Comments on the Appeal to Shame

Joe’s comment begins:

I can be clearer about the origins of the shamefulness of the critic’s error: it has to do with the unwillingness to confront something as threatening as irony, and the corresponding immaturity of preferring a wish (finite irony) to the truth (infinite irony). This does not necessarily imply a strong feeling of contempt on de Man’s part, any more than a teacher or parent must feel strong contempt when imparting a lesson. We do not feel contemptuous of a child who wants a dead parent to still be alive; all the same, it is important that the child eventually accept the truth. It might be shameful for an adult to realize that she is acting like a child in this respect, by succumbing to magical thinking.

Most of this argument is already quite clear in Joe’s previous posts. When Joe characterizes the critic’s error as an “the unwillingness to confront something as threatening as irony, and the corresponding immaturity of preferring a wish (finite irony) to the truth (infinite irony)” he assume that the motivation of this error is immaturity and wish, or rather, assumes that this is de Man’s claim. This assumption is thoroughly negated by my argument that de Man “locate[s] the source of erroneous interpretations not in the desires of a self-motivated subject, but rather to properties inherent in language itself”, which I will defend in response to Joe’s comments below.

The remaining lines of the passage quoted above seem to be offered in response to my attempt to insist upon a distinction between error and shame in the above post:

If, on the one hand, the desire to control irony is genuinely legitimate – if, for example, it is the product of something other than professional self-interest – then it is not grounds for shame, but rather the source of an error to be avoided (unless we’re going to equate error with shame, which verges on transforming all intellectual discussion into a vast battle of egos in which it would be impossible to respectfully disagree, a conclusion which is contra-indicated by the large degree of respectful disagreement which characterizes the back-and-forth between Joe and myself)

Joe’s contribution seems to be to attempt to displace my distinction between error and shame with a distnction between a contemptuous shame and a non-contemptuous shame. This helps account for the fact that de Man doesn’t seem to be so much making an accusation as correcting an error, without conceding the shamefulness of the error.

I don’t think this distinction is legitimate, however, for several reasons. First, Joe is only able to legitimate the distinction by switching from a discussion of the “appeal” to a discussion of the “effect.” It is possible to imagine a non-contemptuous shame only if we focus on the effect of a text and not on its appeal. Consider Joe’s example – a child who wants a dead parent to still be alive. The psychologist (or remaining parent, etc.) attempts to help the child accept the truth, even though this might cause them to be ashamed of their desire. The appeal is non-contemptuous, even though the effect is shame. Still, the appeal is not to shame – the psychologist isn’t trying to make the child feel ashamed so much as to help the child to identify an error which is not worthy of contempt or shame.

Shame is always a potential side effect of recognizing error: Joe is right to say that “it might be shameful for an adult to realize that she is acting like a child … by succumbing to magical thinking” because shame is one possible (though, I would suggest, unlikely) response by Booth. But it would be a mistake to confuse this possible response with the text’s appeal. Not only would doing so amount to committing the affective fallacy (evaluating a text based on its emotional effects on one reader, in this case, the imagined reactions of Booth), it would also mean reducing all critiques to appeals to shame, insofar as it is always possible to be ashamed of having committed an error.

Joe continues:

De Man’s critique of Booth is excessive, and that’s the key here. He doesn’t merely say that Booth is wrong; he tells us that Booth wants to be right and that Booth has a stake in being right.

My response is to deny that this is excessive. As I have argued previously, in order to respond responsibly to Booth’s article, he must tell us that “Booth wants to be right and that Booth has a stake in being right” for reasons that have nothing to do with shame. In order to avoid the inevitable repetition of an error, it’s necessary to understand why the error is repeatedly committed. As I state above, “de Man does have to conjure a critique of professional self-interest in order to ensure that his argument can be evaluated independently of its consequences for the profession”. And in identifying the particular source of the error, de Man is obliged, as a rigorous reader, to attribute such errors to practical necessities, because Booth admits his own motivation in such terms. Again, to quote from my original post – “when de Man says that American critics like Wayne Booth are motivated by practical considerations, he is actually quoting Booth himself.”

2. De Man on De Man

Joe continues:

The rhetoric implicitly excludes de Man, because de Man would not be able to see the matter so clearly if he was a stakeholder possessed by the same desires.

If the reason that de Man is not being patronizing is that language itself produces and encourages these kinds of errors, then we have to wonder why de Man was able to both see through this automatically generated error, and why he would think that an essay of his making could successfully communicate an argument about the undecidability of language. The argument about undecidability (for example, the difficulty of deciding the meaning of the word “legitimate”) is backed up with the claim that de Man “goes out of his way to undermine the very basis of the appeal to shame.” I agree with you about de Man’s intentions here, but the only reason I’m able to accept the claim as true is my non-infinite method.

I see two separate questions in these paragraphs. First, if the erroneous attempt to control finite irony is the product of the properties of language itself, how is it possible to “see through” this error? The answer seems fairly straightforward here, so perhaps I’m missing the point. There is no obvious reason why something produced by language could not be subsequently revealed as an error either by human intervention, or by language itself. In de Man’s account, language both produces the inevitable error and produces the inevitability of its undoing. De Man’s method of reading tends to begin by citing or laying out a hermeneutic account of language (usually some other critic’s reading of a literary or philosophical text) which attempts to control the system of tropes (including irony) in a literary or philosophical text. De Man proceeds by pursuing that reading until it derives a fundamental contradiction. The inevitable attempt to control a tropological system is disrupted by its own impossiblity of completion. De Man is able to see through the attempt to control language because he participates in it, following it to its (il)logical conclusions. The reading of infinite irony always passes through the reading of finite irony. This is one way of understanding the logic of deconstruction – one does not need an outside position of superior knowledge to undo a textual misunderstanding. Language undoes itself when read carefully.

The second question implied by the paragraph above is how can de Man hope to offer a stable critique if his own text insists upon the necessary instability of all language? How can we decide whether de Man is or is not appealing to shame if his method insists on undecidability?

This is a fundamental misunderstanding of de Man’s use of the term undecidability – indeed, it is the misunderstanding I allude to at the beginning of my previous post when I suggest that Joe’s “attempt to confront de Man’s explicit intentions with ‘the totality of what we can say about the way his texts work’ (as opposed to the totality of what we can claim and support about the way his texts work) results in an insufficiently critical hermeneutics which is not, as Joe insists, authorized by de Man’s approach.” De Man’s concept of undecidability does not authorize us to conclude that any and all possible readings of a text are legitimate or true, and in this particular case, it does not authorize us to conclude that de Man is appealing to shame. In other words, it is not necessary to take up what Joe calls a “non-infinite method” to call the “appeal to shame” argument a misreading.

3. Defining Undecidability: De Man on Yeats

To understand why de Man’s notion of undecidability and my own reading of his text as independent of an appeal to shame are not mutually exclusive, it is first necessary to distinguish de Man’s concept of undecidability from a similar concept with which it is often confused:

polysemy – a linguistic term for a word’s capacity to carry two or more distinct meanings, e.g. grave: ‘serious’ or ‘tomb’ … In some modern linguistic and literary theory, it is argued that all signs are polysemic, and the term has been extended to larger units including entire literary works. [from the Oxford Concise Dictionary of Literary Terms]

The concept of polysemy is central for the contemporary theory and praxis of literary interpretation, and involves the fairly well accepted idea that texts and their various subunits (from the chapter down to the word of even the letter) have multiple meanings. Unlike the concept of undecidability, however, polysemic meanings tend to exist in harmony, so that it is possible to choose one or several meanings among many either based on taste (according to reader-response or reception based approaches) or according to pre-established methodological rules such as those Booth was attempting to lay out for irony.

Joe’s argument seems to force a choice between these two models of polysemy: either all possible meanings are legitimate (including the argument that, despite all evidence to the contrary, de Man is calling professionalism shameful rather than merely erroneous when he says, mockingly, that it’s legitimate to want to control irony); or we have to establish some rules for choosing one meaning over another, and thus reduce the text to a singular meaning. It is possible to read the entirety of de Man’s oeuvre, or at least, his notion of undecidability, as an attempt to undo this false choice.

Allow me to explore undecidability through an example from “Semiology and Rhetoric.” In that essay, which appears in Allegories of Reading, de Man discusses the final lines of Yeats’ poem, “Among School Children”:

O chestnut-tree, great-rooted blossomer,
Are you the leaf, the blossom or the bole?
O body swayed to music, O brightening glance,
How can we know the dancer from the dance?

According to de Man, the final line of the poem has two possible meanings – we might be tempted to call it polysemic. Just as the question “What’s the difference?” can mean either “Please tell me what the difference is” or “I don’t care, it doesn’t make a difference” the question “How can we know the dancer from the dance?” can mean either “We can’t know the dancer from the dance” or “Please tell me how can I know the dancer from the dance?” Both questions can be read either interrogatively (as real questions, expecting real answers) or rhetorically, as calling into question whether the real question can even be answered.

In the case of the Yeats poem, the tendency among critics is to read the line as a rhetorical question, and thus to insist upon “the potential unity between form [the dancer] and experience [the dance], between creator [the dancer] and creation [the dance]” (11). But de Man suggests that, even taking account the rest of the poem, the alternative reading is equally likely. The alternative reading, which suggests that there might be some as yet unknown method of telling the dancer from the dance, thus negates the rhetorical question’s assertion of unity between form and experience, creator and creation.

The readings are incompatible – we can’t simply say that the line means both, because they disagree with each other – but there is nothing in the line itself which allows us to decide between one reading and the other. If we try to contextualize the lines in light of the rest of the poem, the same undecidability remains. If we try to contextualize the lines in terms of Yeats’ own life and historical context, the context proves equally ambiguous (admittedly, de Man does not fully support this claim because it is included as a model of his argument rather than a proof); more problematically, this sort of move – referring back to Yeats’ life as the guaranteer of meaning – already involves deciding whether or not we can know the dancer from the dance, the creator (Yeats) from the creation (the poem). De Man presents the challenge for critics as follows:

Neither can we say that the poem simply has two meanings that exist side by side. The two readings have to engage each other in direct confrontation, for the one reading is precisely the error denounced by the other and has to be undone by it. Nor can we in any way make a valid decision as to which of the readings can be given priority over the other; none can exist in the other’s absence (12)

The passage is a direct refutation of a certain version of polysemy. While a Booth-like system of rules for interpretation doesn’t suffice to control the polysemy of the text and reduce it to a single meaning, we also cannot simply choose one meaning over the other, since the poem is defined by the relationship of opposition between its two possible meanings. The poem has not simply accumulated additional registers of meaning, as if one or the other meaning were merely an accident or a coincidence; no, the poem is so thoroughly infused with these two opposing meanings that they must both be necessary – the impossibility of deciding whether we can know the dancer from the dance is indeed the point.

Thus we move from the impossibility of deciding the dancer from the dance to the impossibility of deciding whether we can decide. The infinite irony of the situation is that suddenly de Man’s implicit question – “How can we know if we can know?” which we were apt to read rhetorically (as in “we can’t know if we can know”) suddenly becomes opened up to the possibility of a literal reading (as in “please tell me some method by which I can decide between these two readings of the text” – namely, Booth’s question!). De Man’s text thus contains both the necessity and the irresolvability of questions like Booth’s. It is, in a sense, therefore not immune from it’s own claims of undecidablity.

4. Mistaking Undecidability for Polysemy: De Man on Schlegel’s Lucerne

Polysemy and undecidability are easily confused, and Joe’s original post on de Man’s “The Concept of Irony” furnishes an excellent example when he quotes de Man’s discussion of Schlegel’s Lucerne. Here is the original passage from de Man:

There is in the middle of [Schlegel’s short novel] Lucinde a short chapter called “Eine Reflexion” (A reflection), which reads like a philosophical treatise or argument (using philosophical language which can be identified as that of Fichte), but it doesn’t take a very perverse mind, only a slightly perverse one, to see that what is actually being described is not a philosophical argument at all but is — well, how shall I put it? — a reflection on the very physical questions involved in sexual intercourse. Discourse which seems to be purely philosophical can be read in a double code, and what it really is describing is something which we do not generally consider worthy of philosophical discourse. (168-169)

and here is what Joe has to say about it:

This problem of the “double code” then becomes the basis for de Man’s explication of “parabasis,” meaning the continual interruption of the meaning of the text by another meaning. De Man continues, “You are writing a splendid and coherent philosophical argument but, lo and behold, you are describing sexual intercourse” (181). What of it? Is it really so surprising that the problems “involved in sexual intercourse” bear some relation to other problems of communication? It is not ironic that practicing one’s writing is beneficial, and so is practicing one’s free throw — the two things are simply homologous. It is merely that we are supposed to find Schlegel’s homology between sex and philosophy so embarrassing that philosophical discourse immediately and entirely breaks down. Well, maybe Schlegel is right about the perverse core of philosophy, and maybe he isn’t, but in any case homologies do not mean the end of literary readings.

Joe’s move is essentially to reduce de Man’s reading to polysemy, to suggest that all he has done is to identify two different registers of meaning in Schlegel’s text, two registers of meaning that are at worst superficially incompatible, but that certainly don’t seem to disrupt each other in significant way. If this were the extent of de Man’s argument, there would be little more to say than “What of it?” as Joe says. The mere existence of two ways of reading a passage does not infinite irony make, and certainly not if infinite irony is anything as formidable as de Man makes it out to be. But consider the following passage that falls between where the quote ends on p. 169 and where it “continues” on p. 181:

I’m not going to refer to this, but there is a particular scandal here, one which got Hegel and Kierkegaard and philosophers in general, and other people too, very upset. It threatens in a fundamental way something which goes much deeper than this apparent joke (It is a joke, but we know that jokes are not innocent, and this is certainly not an innocent passage). There seems to be a particular threat emanating from this double relationship in the writing which is not just a double code. It’s not just that there is a philosophical code and then another code describing sexual activities. These two codes are radically incompatible with each other. They interrupt, they disrupt, each other in such a fundamental way that this very possibility of disruption represents a threat to all assumptions one has about what a text should be.

De Man is stating a claim not only that this double meaning is a threat, but that it is a particular threat, that the threat is something more than “just a double code.” Unfortunately, he merely alludes to such a threat and never tells us what exactly, in particular, it is. We get the joke, in that we’re told there is a joke there, but we’re never really allowed to “get” the joke, in that we never understand what Schlegel is laughing at, at the expense of all of these philosphers. He says he’s not going to “refer” to it, which is a shame, because he’s already refered to it, and piqued our curiousity. If I were a more committed scholar (or, rather, less committed to other obligations, with more time on my hands), I would go learn German, read Schlegel’s “Eine Reflexion” in the original language, then go read Hegel and Kierkegaard to figure out what got them so ticked off, and come back, and let you in on this little joke that probably isn’t all that funny, especially since de Man seems to take it so seriously. But I can think of very few things I have less of a desire to do.

The point, however, is that there is more to the Lucern joke than the mere existence of a double code. We can fault de Man here for not telling us what, though he is partially to be excused by the fact that the essay is taken from a preliminary lecture, delivered out loud in a context in which perhaps it might not have even been appropriate to tell the joke, never completed revised into the final form de Man had set out for his essay on Schlegel and Fichte, and published only posthumously. Regardless, even though the joke is not told, we are told that there is more too it than mere polysemy. In order to qualify as genuine undecidability or infinite irony, there must be a direct and necessary engagement between the two possible meanings that accumulate onto a text. Otherwise, it remains entirely possible to say “What of it?” and dismiss one of the meanings as mere coincidence.

5. De Man on De Man continued: Deciding on the Appeal

Thus we are finally in a position to understand how de Man is able to articulate a notion of undecidability without forcing our own decision about the meaning of the line “legitimate to want … to control … irony.” De Man’s method insists on methodological rigor, which means pursuing the attempt to control the tropological system as far as possible. This means that we can use context to resolve variants on meaning as far as possible. De Man does not banish critical reading in the interest of a watered down version of the free play of signifiers. He doesn’t allow any reading possible, and in fact spends most of his essays critiquing particular misreadings. It thus remains entirely possible, without stepping outside of de Man’s claims to infinite irony, to denounce a misreading which is undercut by contradictory evidence. The point of undecidability is not that this will never be possible, but rather that it will sometimes be impossible, or, in a slightly stronger version, every attempt to construct a stable logical system will be undercut by such undecidability at some point.

As observed above, de Man is no exception. His own text is subject to the same undecidability insofar as it is condemned to express the necessity of the very methodological rigor it declares impossible to carry out.

6. The Ethics of Truth: a Category Mistake?

Joe continues:

Let’s assume, for the moment, that the reason de Man is able to correct Booth and Rousseau has to do with giving up his stake in the finitude of irony; that is, his preference for a disinterested account of the real truth over an interested distortion.

What can we say about this sacrifice? Obviously, we cannot justify de Man on the grounds that in the long run, texts will prove too unstable for the uses to which we want to put them. That would be the worst kind of speculative appeal to consequences wedded to generalization, and would furthermore immediately posit an “outside” to the text that would act upon us to make us sorry for our limited readings.

The alternative is an appeal to “truth for truth’s sake,” without concern for consequences, and curiously enough without even the possibility of consequences, a lack that separates this kind of philosophical drive from the unknowable consequentiality of non-applied science.

“Truth for truth’s sake” in this sense is the oldest appeal to philosophical integrity in the Western tradition: it is a direct reference to the character of Socrates, as he appears in the dialogues of Plato. Socrates equates the good with the true, and asserts that all people are compelled by the truth insofar as they understand it. Socrates defends this proposition against the charge that goodness and truth lead to unhappiness, and thus ultimately links goodness, to truth, to pleasure.

One could certainly uncouple de Man from Socrates, though this strikes me as historically irresponsible, just as one could take apart his argument in modular fashion. De Man can still be making a truth-claim, regardless of whether he’s making an appeal to shame, as you suggest. The same is basically true of art, particularly written work. One can imagine a Fitzgerald novel in which Nick Carraway’s disillusion takes place without the accompaniment of Gatsby’s death.

In my view, taking de Man’s argument apart in this fashion is actually highly reductive of de Man, because it means condensing his essay down to aphoristic articulations of critical difficulties (such as “undecidability”). Presumably, if such a dismantlement and condensation were desirable, de Man himself would have taken the lead.

In addition, the point of rhetoricizing de Man’s appeal to the truth is actually to legitimate de Man rhetorically; otherwise, de Man is in the position of arguing for our responsibility to truth, without explaining where that responsibility comes from. You may reject my idea that he is proposing a responsibility to true Pleasure (simultaneous in Plato with the responsibility to the Good), but unless he can come up with new, non-Socratic grounds, he is simply leaning on a tradition (i.e. the Western philosophical tradition) that has always cast itself as a protector of man’s best interests, both psychologically (happiness or pleasure) and physically (the well-ordered society, the prospect of living in harmony with metaphysical laws).

I’m not exactly sure how to respond here. The question here is not about the possibility of a true discourse, but about the ethics of truth or the appeal of truth. I can’t really say why truth is appealing, and I can’t say why de Man’s arguments are appealing. I do admittedly suggest briefly in my previous post that de Man’s argument is appealing because it is true, that it appeals to impersonal truth. I’m not sure if this is true or strategic. I do think it’s a mistake to identify the appeal of de Man’s argument with bliss or shame, and I think I’ve clearly identified the reasons why this is a mistake. But I honestly can’t say why I find de Man’s arguments appealing other than that I really think they’re true. I’ve yet to read an argument that systematically critiques them without falling into the errors de Man has already anticipated.

In the end, I think this question about truth for truth’s sake vs. for the sake of pleasure or goodness is a category mistake. I think I would be content if this argument ended with the conclusion that my arguments were true, whether or not this truth could be affirmed as valuable. I feel like this would suffice to undermine arguments for finite irony. After all, I readily acknowledge that it may be necessary to say that irony can be controlled, even if it can’t be. It is no less true that it can’t be controlled, and that it is necessary to say this as well. Similarly, I feel like in an argument about the interpretation of a particular text, it would suffice to show where an erroneous argument diverged from the text. I doubt the interpreter would go on defending his or her reading because of its utility, even if I can’t say why he or she shouldn’t. And if he or she were to go on defending his or her reading in the face of conclusive evidence to the contrary, I would have to understand this insistence on the claim as disingenuous, i.e. as ironic, thus merely confirming the inevitability of infinite irony.

This is an insufficient response, and I need to think more about it. In part, my answer will have to accompany my response to the second half of Joe’s previous post, the response I promised to postpone until after the discussion of the first half of his post was resolved. De Man’s third essay on Nietzsche (“The Rhetoric of Persuasion”) will go a long way towards clearing up this problem – Nietzsche is definitely the key here. “On Truth and Lie in the Extra-Moral Sense” may also be relevant. In the mean time, Joe, I would appreciate an elaboration of this point – why does it matter what the appeal of truth is? If de Man does indeed lack a convincing defense of truth, does it follow that his arguments are less true? How do you respond to them in the absence of a defense of truth as a value? At what point in this debate does de Man’s failure to identify “for what’s sake” he speaks the truth become a problem?

I’ll end, for consistency’s sake, by quoting the remaining lines of Joe’s comment, with which I have no quarrel:

As for the affective fallacy, don’t confuse my descriptive accounts of nihilistic irony with a personal psychological complex. I have a lot of sympathy for the desire to be radical, even when it doesn’t come accompanied by a workable alternative, if it is contextualized as a specific moment of joy or misery. When the desire for the “break” appears in a work of art, it is a moment rather than an abstract truth (of fragmentation, of unresolvable antagonism).

Furthermore, infinite irony is entirely achievable as a hysterical stance: hence Stephen Dedalus and the young man in Kierkegaard. My newest post on Buffy the Vampire Slayer, as well as earlier posts on other pieces of pop culture, are based on reading irony as a continual and continually unsuccessful flight from a suffocating mass of dangerous or dispiriting fictions. It’s my position that the failure of these rebellions (also part of the works in question, where the ironists are prodigal daughters and sons) has something to do with how irony is being applied. The makeup of modern irony is described very well, and valorized, by de Man, which may explain his continual usefulness to literary critics.

Advertisements

~ by Matt on March 4, 2007.

One Response to “Irony Continued: Undecidability and Polysemy, Truth and Error”

  1. […] stake in the ’stable versus unstable’ irony issue.  There’s one final follow-up here.  I’d urge you to read the whole exchange; you’ll learn a lot about the way ‘irony’ has […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

 
%d bloggers like this: