<< There Is No Such Thing As Intelligence | Front Page | Dretske on 'a curious blend of the conventional and the natural' >>
Sunday, June 17, 2007
We need to interpret things, as well as interpretations
E.D. Hirsch’s famous sentence is ‘My car ran out of gas,’ which is supposed to be ambiguous - indeterminate - between a ‘ran out of fuel’ interpretation and a ‘my Pullman dashed out of a cloud’ interpretation. The idea is supposed to be that no ‘piece of language’ could ever have a determinate meaning unless it were nailed down with reference to authorial intentions. This is a classic example, for inaugurating a somewhat hoary dispute between ‘formalists’ and ‘intentionalists’. But that’s not what I really want to ask about. What I want to ask about is: what contributor to this debate focuses on how interpretation of these sorts of texts is to a considerable extent a matter neither of thinking about what words mean, nor what people intend by their words, but about the implications of situations?
Couple example sentences (which I am borrowing from an old paper by John Haugeland, who is talking about this thing I think more people should really consider):
1) The box was in the pen.
2) I left my raincoat in the bathtub, because it was still wet.
3) Though his blouse draped stylishly, the toreador’s pants seemed painted on.
It seems obvious that interpretation of these sentences proceeds largely by imagining potential ‘situations’ - you have to be familiar with the properties of objects, which you hypothetically manipulate in your mind, trying different interpretations by moving different props around, and altering them experimentally.
Running down the list, regarding 1) there might be a big wooden crate in a pigpen, or a tiny sealed compartment in a spy pen, containing microfilm. In large part you work out the possibilities not by thinking about what words mean, or what anyone is likely to have meant, but by sorting ranges of actual physical possibilities. For example, a crate wouldn’t fit into a writing pen. That’s not a truth about English or a truth about what anybody intends. But its a fairly relevant constraint. 2) is even better, because, in order to figure out that ‘it’ refers to the raincoat and not the bathtub, you run through a fairly obvious path of practical reasoning. This doesn’t really have anything to do with the meanings of English words, or with hypotheses about what someone means by the sentence. Rather, you think about what sorts of activities a person would be likely to engage in. (If you wanted to get questions about the speaker’s intentions clean out, change it to ‘he left his raincoat ...’ 3) It is more likely for a toreador’s pants to be tight than paint-spattered. You might say that this is probably what the speaker means, but that judgment about likely intention is actually an inference from prior knowledge about the typical natures of toreador pants.
Well, I don’t really have the energy to develop this thought tonight, but I’m interested in it because it seems like a subject that is actually rather neglected in discussion of literary interpretation - literary meaning. The ‘meaning’ of a text, i.e. the sorts of things interpretations of texts aim to extract, is largely a function of the implications of situations that arise in (say) a story. You interpret the text by interpreting situations, which is largely a matter of appreciating what situations imply. But the implications of situations are neither a function of the meanings of words, or the intentions of authors.
I could put it a different way by asking: who has noticed that literary meaning is, to a considerable extent, a function of what H.P. Grice called ‘natural’ (as opposed to ‘nonnnatural’) meaning. That is, relations like clouds mean rain, these spots mean measles, pheasant tracks mean pheasants? (I sort of asked a version of this question a few months ago, but I’m asking again. Because I’m still curious.) Example: the opening of Macbeth. ‘When shall we three meet again, in thunder, lightning or in rain?’ You need to be able to understand that this implies the ladies are a bit weird, or up to something weird, because they are planning to get wet and uncomfortable, apparently. That the ladies are planning to meet in uncomfortable weather is not something extractable from the meanings of the English words they use. You could say that Shakespeare intends us to notice they are weird. But really this is an effect of the inference patterns I am drawing attention to, not an explanation of how they work. You figure out that the ladies are weird just the way you would figure out that a real lady, standing in real rain for no apparent reason, was weird: she’s standing in the rain, getting wet. Why is she doing that? So you reason about characters the way you reason about people. You reason about fictional situations the way you would reason about real situations - to a considerable extent. (Obviously there are limits to this. My point isn’t that literary meaning is ‘just’ natural meaning. Rather, that theorists tend to discuss literary meaning as though it were all either linguistic or intentional. There isn’t any role for the the meaning of things, putting it baldly. But there should be.
I’m interested in critiquing theories of interpretation that overly linguistify or intentionalize meaning. They fail to take account of the degree to which what needs interpreting are things, not interpretations (would be a slogan of sorts.)
Comments
This kind of thing is one reason why I’m more interested in literary science fiction than in generic literature. There is at least the potential that one’s ability to interpret situations will be affected by the appearance in these situations of objects, social modes, history, ideas etc. that do not exist.
I quite agree: science fiction is the genre that most perfectly illustrates the situational nature of most literary interpretation. The significance of what is happening in a SF story is largely parasitic on the horizon of assumptions about what is or is not be possible in our own world--why SF often requires technical training to appreciate. Also, the interpretation of SF neologisms depends quite specifically on situational context. Take S:
S: Leaving the bustle of Manhattan behind him, Chuck stepped through the X and finds himself in Paris.
The meaning of X--and S--is decided through an interpretation of Chuck’s situation, not the interpretation of sentences. Then again, W.B. Michaels would say that the intention of S’s author is to evoke a mental model of a situation in which X is some sort of teleporter, so situational interpretation really just leads us back to interpreting intentions.
John—The sort of interpretation you’re talking about has received considerable attention in the natural language processing corner of cognitive science. Back in the 70s researchers quickly discovered that, in order to get a computer to “understand” even a simple text—where it, say, demonstrates this understanding by answering questions—you had to program it with considerable situational knowledge in order to make the kinds of inferences you’re talking about. This is sometimes known as the “common sense” problem; understanding language depends on an enormous store of relatively trivial commonsense knowledge to be used as the basis of such inference.
Back in 1978 I filed a dissertation on “Cognitive Science and Literary Theory” in which I devoted a great deal of time and attention to such models and the type of inference they automatially entail, though I didn’t dwell all that much on such inference as a big deal, mainly because it simply goes with the territory. When you’re in that world, you simply assume that inference; that’s why you create the model. In 1976 I published an analysis of Shakespeare’s sonnet 129 (PDF) based on that work.
You might want to ask Jonathan Goodwin whether or not David Herman’s recent Story Logic goes into this, as Herman is interested in reconstructing narratology through cognitive science and Goodwin has reviewed the book.
"interpretation of these sorts of texts is to a considerable extent a matter neither of thinking ... about the implications of situations”
Are you using the periphrasis “about the implications of situations” rather than the more obvious “about context” out of sensitivity to the feelings of SEK, who has recently suffered so much unjust unjoy after speaking intelligently upon that subject?
Actually, the more “natural” literary interpretation of that opening trope by the “weird sisters” in “MacBeth” is that they are being identified with the occult “powers” of nature, rather than simply being identified as “weird”. This is because the peripety upon which the “tragedy” of MacBeth ends up turning is that the “weird sisters’” oracle turns out to come about through entirely naturalistic means, such that it is MacBeth’s superstitious credulity combined with his illicit ambition that ends up bringing about his downfall. MacBeth doesn’t have a fool, because he is his own fool, and he has no players because he himself is a “player”. He’s a natural-born liegeman, who transgresses his own dread/awe/reverence of sovereignty. The hierarchical “point” of the “tragedy” reconciles with its “enlightening” irony.
So if the point here is some sort of reader response approach, that might be valid and worthwhile. But the point seems to be missing here that literary works “work” through a disruption of meaning-expectations. And that’s because literary works don’t “mean” formally, or semantically, or intentionally, or situationally, or contextually. They “mean” modally, which is part of what ‘defines” them as “literary” and part of their “point”. And why they tend to disrupt meaning-expectations.
Zizek’s discussion of Wittgenstein in the fourth chapter of For they know not may be relevant. (It’s the non-"Zizek book” Zizek book! The constitutive exception of his body of work!)
john c halasz, I think you are missing the point. Which is that - well, of course it’s complicated. If you want to say that literary works ‘mean’ modally, then fine. But they also work in part because we apply reasoning patterns to them that just import straight from the ‘real world’. Bad weather means discomfort. This isn’t the key to understanding. But it is an unacknowledged element - much of the time.
Bill B. mentions the cog sci literature. That is, indeed, the place where I have found it discussed. I should check out ‘story logic’, I guess.
Adam, I was avoiding ‘context’ because the word tends to be ‘textualized’ more than situation. Situation is a thingly word, and I’m looking for that at the moment.
John writes, “I could put it a different way by asking: who has noticed that literary meaning is, to a considerable extent, a function of what H.P. Grice called ‘natural’ (as opposed to ‘nonnnatural’) meaning. That is, relations like clouds mean rain, these spots mean measles, pheasant tracks mean pheasants?”
Isn’t that the whole point of Derrida’s discussion of Husserl on expression versus indication? Our speech “expresses” an intended meaning; our blush indicates our embarrassment. Derrida, of course, deconstructs this binary, leaving us with an indicative foundation for all expression: the trace, which indicates the impossibility of ever recovering the full presence of intention.
Literary genre seems to work indicatively. That is to say, just as smoke indicates fire, certain figures, tropes, images, plot moves, etc. indicate a certain body of associated conventions being called into play. So even if a novelist doesn’t intend to bring to mind *every* gothic device, her use of a gothic figure nevertheless conjures (indicates) the entire history of gothic conventions.
Luther, where exactly does Derrida discuss Husserl on indication? I could actually use a citaton for that.
Also, it’s important to distinguish natural from nonnatural indication, I think (it’s the natural/nonnatural axis that is crucial, not the analysis in terms of indication.) Literary genres working indicatively would be a case of implicature, according to Grice. I’m saying there needs to be more emphasis on natural meaning (not that I’m so absolutely sure about this, or that I am so sure we need to use Grice’s term, which might just confuse people in various ways.)
Count me as a vote in favor of “might just confuse people.”
Hey John, I think I mentioned this book a while back, but it’s relevant again:
Horst Ruthrof (2000) The Body in Language (London: Cassell).
Ruthrof puts forward an extended argument about meaning as being always and necessarily “heterosemiotic”, by which he means that “meaning” of any kind requires the exchange of signs from one semiotic system for (or into) signs from another semiotic system. These signs include what we might call “phenomenal” or “corporeal” signs, i.e. perceptions (though don’t invest too much significance in those words since they are mine; I don’t recall whether Ruthrof uses precisely those terms). In other words, he’s talking about what you’re calling “natural meaning”.
Ruthrof once gave me as an example the following rather idiosyncratic colloquialism, which struck him (being a Bavarian in Australia) as rather bizarre but instantly meaningful the first time he heard it: “I’m so hungry I could eat the crotch out of a low-flying duck”. I can’t remember if he uses that example in his book, but I’m sure he’d use something similar.
Basically, Ruthrof’s point was that someone who’d never come across the phrase before would have to visualise the event (or “situation”, to use your preferred term) in order to get the drift of the phrase. They would have to imagine a low-flying duck and someone trying to take a bite out of the duck as it flew overhead and thus recognise how hungry you would have to be to try doing that!
Going from this and many of your other posts, I think you’re trying to work towards the same kind of position and arguments that Ruthrof’s put forward in The Body in Language and in his earlier work, Semantics and the Body: Meaning from Frege to the Postmodern.
John — Derrida on Husserl and indication: Speech and Phenomena.
1. I want to echo the recommendation of others of Speech and Phenomena which is both wonderful and one of Derrida’s texts that can be taken in a more Analytic sense.
2. It’s possible that I mentioned this last time you brought this up (it all seems vaguely familiar) but Gadamer specifically addresses the interpretation of texts in (what else?) Text and Interpretation. Though it is a good read it might be not specialized enough for your purposes after all it is quite short and more concerned with setting himself off against Derrida than actually explicating a theory of textual interpretation. Nevertheless I highly recommend it. (And it is oddly apropos of the discussion in the other thread)
Also, it’s important to distinguish natural from nonnatural indication, I think (it’s the natural/nonnatural axis that is crucial, not the analysis in terms of indication.)
I just got even more confused than I already was, perhaps because I’ve been assimilating natural meaning to Blue Book–style criteria & to a lesser extent Cassirerean signs (as opposed to symbols). Here is why I was already confused: these spots mean measles not because they are some sign over and above measles pointing at measles-having, they are part and parcel of measles-having; a part that happens to be easily recognizable and relatively enduring. Similarly if things are working correctly procedurally, that the budget looks like this is just part of the firm’s financial situation being the way it is. But it’s in no way part of the bus’s being full that the driver made these three toots of the horn, or whatever the example is; there, there is a gap (& this, I take it, is what lets something other than that the bus is full be meant by the toots on the horn). That is to say, I don’t really understand the claim that these spots mean measles as having anything to do with what knowledge one might have in virtue of which one would be able to make the claim “these spots mean measles”.
Tho’ I could see claims (especially in the context of, say, imaginative resistance about values) along the lines of, if such-and-such happened (in the story) then such-and-such else must be the case. But that doesn’t seem to be an issue about literary interpretation. You would say the same thing in “real life” if someone told you about his day, or if you witnessed something.