Monday, July 25, 2005
Near Theory (My creation, is it real?)
Little game. Get out your pencil. Provide the following (or just click and find out what I'm on about under the fold.)
1. Singular noun. Two syllables.
2. Verb. One syllable.
3. Plural noun. One syllable.
4. Singular noun. One syllable.
5. Singular noun. One syllable.
6. Plural noun. Two syllables. Must rhyme with the selection for 1.
7. Singular noun, two syllables.
8. Singular noun, one syllable.
9. Verb, one syllable.
10. Verb, one syllable.
11. Adjective, three syllables.
12. Plural noun, one syllable.
13. Plural noun, one syllable.
14. Plural noun, one syllable. Must rhyme with the selection for 10.
Following up on Sean's post, I am inflicting on Knapp and Michael's "Against Theory" a refutation by mad lib. I ran this little experiment on the wife as test subject; here is the product.
An ashtray did my spirit seal;
I rode no human shoes;
She seemed a vase that could not feel
The tent of earthly news.
No anthill has she now, no horse;
She neither knits nor toasts.
Rolled round in earth’s bejeweléd course,
With frogs, and blots, and ghosts.
As you can see, we are punching out bits of the Wordsworth poem whose interpretation is - by tradition - the bone of contention in these sorts of arguments; we're slotting in our selections. As you can guess, the form of the refutation is, basically: no author, no authorial intention. But plainly the thing is meaningful; you could interpret it. (We shall.) So Knapp and Michael's thesis that meaning simply is authorial intention must be false. The obvious response would be: but if you interpret it you are assuming an 'as if' stance, positing an author. This doesn't work, but it will take a while explaining why not. (However, for your convenience, I've made that punchline - a partial interpretation of my poem - into a separate post for those who can't bear to read all that appears below.
Here's the original, with the variable bits underlined to facilitate your private performance of the above exercise. (Please feel free to show us your results in comments.)
A slumber did my spirit seal;
I had no human fears;
She seemed a thing that could not feel
The touch of earthly years.
No motion has she now, no force;
She neither hears nor sees;
Rolled round in earth’s diurnal course,
With rocks, and stones, and trees.
Anyone want to write a script? Pencil and paper is lo-tech, sitting at home of a Sunday afternoon when there are courageous pioneers out there on the bleeding edge of near science! (You really must watch the video of three guys delivering their randomly generated papers.) Favorite lines:
We need epistemologies that will be symbiotic, flexible, and also operate on a large scale.
Assumption: there are only child-like adversaries.
Extremely mobile information underwater
... And so it prevents the world wide web!
And then there's this. That's all the jokes I've got. Unless you are interested in philosophy of language, this is your exit; go amuse yourself somewhere else. I believe this is the longest blog post anyone has ever written. I think it's pretty clear, but still ... It's a paper draft, an alternative explication of points touched on in my Nabokov/chess piece. Here's a PDF version if you prefer not to scroll endlessly. [UPDATE: portions of this post edited for clarity after posting, and PDF corrected accordingly.]
You can read the first five pages of Knapp and Michaels' original essay yourself, in Against Theory. Amazon is handy sometimes. Read 'excerpt'. Then, if you want to read three more pages, 'search inside' for a distinctive phrase - oh, say 'genius of the shore', which will get a hit on the next page; then go next page, next page. And that's all Amazon will allow. But the first five pages contain the thesis, relevant definitions, and full statement of the argument to the conclusion that, necessarily, meaning just IS authorial intention; that "The meaning of a text is simply identical to the author’s intended meaning" (p. 12).
The argument is a thought-experiment. Fish's argument in his NY Times editorial alters nothing but incidental stage-dressing. See Sean's post for discussion and lengthening commment thread. (Matthew Yglesias offers a refutation similar to my mad lib strategy.) I'll quote a long bit before the NYT firewall is kindled:
Suppose you're looking at a rock formation and see in it what seems to be the word "help." You look more closely and decide that, no, what you are seeing is an effect of erosion, random marks that just happen to resemble an English word. The moment you decide that nature caused the effect, you will have lost all interest in interpreting the formation, because you no longer believe that it has been produced intentionally, and therefore you no longer believe that it's a word, a bearer of meaning.
It may look like a word - it may even seem to be more regularly formed as such than the scratchings of someone who is lost - but in the absence of the assumption that what you're looking at is a vehicle of an intention, you will not regard it as language. It is not until you change your mind and become convinced that the formation was, in fact, designed, that the marks will become language and it will be appropriate to interpret them.
Even then you are not home free; just because you're now sure that the marks spell the word "help," you still don't know what it means. It could be a message from a person in distress. It could be a direction like those on a computer screen ("Need help? Look here."). It could be a petition to God. It could be a reference to a Beatles song. Scrutinizing the word won't tell you which of these things it means.
This is why Justice Scalia has it backwards: if you're not looking for what is meant, the notion of something being said or written is incoherent. Intention is not something added to language; it is what must already be assumed if what are otherwise mere physical phenomena (rocks or scratch marks) are to be experienced as language. Intention comes first; language, and with it the possibility of meaning, second. And this means that there can be no "textualist" method, because there is no object - no text without writerly intention - to which would-be textualists could be faithful.
Knapp and Michaels do the same work by imagining successive stanzas of Wordsworth - what seems like Wordsworth - washing up on the beach. At first you see words in the sand - what seems like words - and you assume someone scratched them there. Then a wave washes it away and replaces it with a fresh stanza (excuse me: pseudostanza.) What is causing it? Speculate away! But (allegedly) all explanations fall into two piles. "You will either be ascribing these marks to some agent capable of intentions (the living sea, the haunting Wordsworth, etc.) or you will count them as nonintentional effects of mechanical processes (erosion, percolation, etc.). But in the second case - where the marks now seem to be accidents - will they still seem to be words? Clearly not. They will merely seem to resemble words" (p. 16).
This argument fails, right down to the awkward formulation of the conclusion. (Surely we can grant the scratchings really resemble words, since this is actually stipulated.)
I think Knapp and Michaels have been decisively refuted by George Wilson [don't have it handy] and by Searle's "Literary Theory and its Discontents," which appears in Theory's Empire. As Sean mentions, my Nabokov/chess piece contains another refutation. This post is, in effect, another angle. One may as well also note: what seems like the correct, obvious objection turns out to be really correct. Why should I say these things in the sand 'clearly' aren't words? It isn't clear at all. Yes, it's weird to say 'the sea is writing words.' But saying they aren't words won't make it less weird what the sea is doing. But Knapp and Michaels' argument hinges crucially on intuition and usage being 'clearly', unambiguously, in their favor.
Still, a failed argument isn't all bad, so long as it teaches. Mostly I'm not going to be knocking them down but kicking off from them when they're down to end up somewhere interesting. And, let it be said, a big reason why the refutations are so decisive in this case is that Knapp and Michaels' argument is absolutely clearly formulated. No room for wriggling off the hook, although two failed wriggles have been attempted to date.
Let's spell out my refutation-by-mad-lib. No meaning for any line of the poem you see above, "An ashtray did my spirit seal," was intended by any intentional agent. Certainly no meaning was intended for the poem as a whole. Wordsworth didn't, I didn't, and my lovely wife, Belle, didn't. That exhausts our slate of candidates. The poem has no author and so, by Knapp and Michaels' argument, is no poem. Indeed, it is not made of words. Only of word-resembling pixel patterns. So if I can convince even of so much as this - that thing you will see by scrolling up is made of words - Knapp and Michaels are refuted. But surely it's made of words because I asked my wife for words. Combining them can hardly have annihilated their natures.
What will Knapp and Michaels reply? I think they will say that, to the contrary, there is a great deal of intention here. The 'poem' is like a parody of Wordsworth, so there are blots and ghosts of the original author's intentions rolling around. I, John Holbo, have intentions: I didn't do this by accident; I intend to refute Knapp and Michaels. I also intend that the result of my very deliberate and stagy exercise should be interpreted. Belle has intentions. She intends certain words. Her stipulative acts make the poem come out as it does.
Does this save Knapp and Michaels? No. All this only goes to show is that there are different senses of 'intention'. When we sort them, we find the plurality prevents the running of the argument. Without further ado, three senses of intention:
1. Intentionality. A technical term denoting anything that exhibits aboutness, be it a thought, sentence, proposition, anything. Searle employs the term in this sense, usually capitalizing, to make clear the usage is non-standard. (Brentano first breathed modern life into this medieval term of art. Here's a nice article.) If you omit the capital, but employ the term in this sense, you get something like 'meaning is intention' as a tautology. We need to keep this in mind, because Knapp and Michaels may skim illicit plausibility off it. We need to scrape that off, because none of the anti-intentionalist views they are denying are attempts to deny that meaning and aboutness go together.
A point of possible confusion. From the linked article: "As the latin etymology of ‘intentionality’ indicates, the relevant idea of directedness or tension (an English word which derives from the Latin verb tendere) arises from pointing towards or attending to some target." This is significant because the New Critics, the usual anti-intentionalist suspects, favor a rhetoric of hermetic textual autonomy - 'poem should not mean but be' - which may suggest a quite absurd refusal to admit that poems exhibit Intentionality. Obviously the denial that poems mean, i.e. are linguistic - as if poetry were some sort of 'absolute music' - is absurd. This New Critical stance is a merely rhetorical gesture, or just a brush off to those they like. When the New Criticism was ascendant, this sort of bullying had effects; when it declines, this extreme rhetoric gets cited to produce an erroneous sense that somehow 'meaning is intentional' - in this basic sense - is something anyone has seriously denied. We shouldn't bother pretending to take the rhetoric literally.
Another point of confusion: Knapp and Michaels quote Searle, "there is no getting away from intentionality," by way of hinting he is just being timid in not pushing his view to its logical conclusion, i.e. their conclusion. But what Searle is getting at is: everything intentional, i.e. about anything, is either mental, e.g. thoughts, or a derivative product of willing actors who have minds. Symbol systems are conventional, not natural; conventions are established by creatures with minds and wills. This primacy of mental agency may seem to get us all the way to Knapp and Michaels; they hint so; but it is not so. The dispute between intentionalists and formalists isn't a dispute between those who think languages are man-made (made by beings with minds and wills) and poor deluded fools who think languages grow on trees or bubble out of the ground. The dispute isn't whether languages need to be made, i.e. whether there is always some intentional act at the root, but whether every subsequent use of language also needs a constitutive intentional contribution to sustain it. A theological analogy: all Christians believe that God created the universe. Occasionally, Christians have been occasionalists; occasionalists believe God's successive acts of will maintain the universe, moment by moment. Intentionalists like Knapp and Michaels (and Fish) are, in effect, occasionalists about language. This is a non-trivial bit of metaphysical speculation. More theology below.
2. Intent. 'Intent' means plan. (Obviously the sense of directedness towards something is operative here.) My intent, in producing "An ashtray did my spirit seal," is to refute Knapp and Michaels, also to facilitate making other points. Although this plan is ultimately communicative, many plans are nothing of the sort. It is my intent to eat dinner. This does not turn me into a symbol for dinner eating. (Then again, people planning to eat dinner are good things to feature in restaurant ads.) In my Nabokov/chess piece I talk about moves in chess. You shouldn't move without having a plan, although you can. But, by formulating a plan and executing it, I am not telling my opponent anything, in a semantic sense. The clearest way to see this is to imagine non-standards case in which chess moves might convey semantic content. Sherlock might establish a code. When he moves his king for the third time, Watson is to go to the drawing room. Or perhaps I am losing and move my king right out in the open so as to say: 'I don't care anymore. I quit.' Mostly this isn't how it goes: moves don't send messages. (You can investigate the meaning of any move, of course, even if no one is sending you messages via it; but that just shows we have too many senses of 'meaning' floating around. More below.)
Wimsatt and Beardsley's 'intentional fallacy' seeks to rule out 'author's intents' in this sense of plan: intentions to do, or realize certain effects. This is rather sneaky, as the most plausible sense of 'intention', qua candidate for being the meaning of a poem, is the one that comes next.
3. Authorial intention. I want to subdivide this category promptly into Gricean and Hirschian piles. H.P. Grice and E.D. Hirsch have highly contrastive - not necessarily inconsistent - intentionalist theories of meaning. (Not that these theorists are necessarily the best, but they are representative enough.)
Grice's notion of conversational implicature. Normally a bad move in chess is not a way of saying something. That changes when the opponent can run through a plausible line of thought, like so: 'he can't fail to see it is a terrible move, and also see that I can't fail to see it is a terrible move, and that I must see he sees that I see this; there is no point to such a move unless he is telling me something ...' Clear enough, apart from some daunting complexities elided on the way to 'I quit!', as conversationally implicated via mutual recognitions of intents to communicate x by saying/doing y.
Hirsch's notion of an intentional horizon. An intentional horizon marks the limits of an author's - or anyone's - awareness, consciousness (and perhaps subconsciousness.)The normative proposal is that, in interpreting a poem, one should stay within this horizon. "The probability of an interpreter's inference may be judged by two criteria alone - the accuracy with which he has sensed the horizon as a whole and the typicality of such a meaning within such a whole. Insofar as the inference meets these criteria it is truly an explication of textual meaning. It simply renders explicit that which was, consciously or unconsciously, in the author's intention" (Validity in Interpretation, p. 223).
It would be interesting to make a study of how Grice and Hirsch might complement and/or contradict each other. Hirsch says little about intents - i.e. plans that x be recognized as an attempt to communicate y. (It is not clear authors intend to communicate their 'intentional horizons' to audiences.) Grice deals well with cases that are, in effect, solvable riddles: tidy allegory, proverbs, metaphors that are poetic packages for messages. He may deal less well with more wide-open terrain - a whole poem that, plausibly, isn't an attempt to deliver a discrete propositional packet according to a protocol of mutually recognized intents. On the other hand, the two views may mesh. I won't try to settle that. Let's just stipulate that 'authorial intention' denotes content - real thoughts in the real mind of a real author - that constitutes the semantic content that an interpretation of a text should (or must) find. (I am strictly reserving 'intent' for plan and 'intention' for candidate meanings.)
I return to Knapp and Michaels and the proposal that my mad lib doesn't refute intentionalism because it is fairly thick with intentionality. The problem is it's the wrong sort. We have Intentionality, but that just begs the question if it is regarded as resolving the question; there is primary intentionality, for example in Belle's stipulations of words. We have intents: my plan to refute Knapp and Michaels; no doubt Wordsworth had plans of his own. But I trust it is clear that the textual meaning of "An ashtray did my spirit seal" - the meaning a close reading of the poem will produce, if any - is not: John Holbo wants to refute Knapp and Michaels; nor, Belle picks 'ashtray'. These plans and acts of stipulation are not candidate meanings of the proper sort. What we do not have here is any 'authorial intention', per sense 3 above. Because, per Hirsch, there is no intentional horizon; the semantic labor is distributed with no one conscious agent having access to the big picture, which is just what the horizon is. Per Grice, no one is in a position to have planned or intentionally determined that, by x, y is meant, and that the audience should get to y by recognizing the plan. For example, no one can have planned that by 'an ashtray did my spirit seal,' the reader is to come to recognize the author's intent to communicate an anti-smoking message, or anything of the sort.
But even by proposing that (rather obvious) reading, am I starting to construct a way out for the intentionalist? Knapp and Michaels can fall back to the position that one can always take up an 'as if' stance. Even if one knows there is no author, one can posit ... an author with an anti-smoking message. And take it from there. Here are Knapp and Michaels, describing the scene by the shore before and after you realize no one wrote the words (or non-words): "You had, without realizing it, already posited an author ... Only now, when positing an author seems impossible, do you genuinely imagine the marks as authorless. But to deprive them of an author is to convert them into accidental likenesses of language.They are not, after all, an example of intentionless meaning; as soon as they become intentionless they become meaningless as well" (p. 16). Presumably one could recover the meaning by the simple expedient of positing again. In this way my mad lib is deflected as a counter-example to the thesis.
This doesn't work. Explaining why not will be my final point (see part II) for which I will need more resources.
First, please note how my distinction between types of intention undoes the neat either/or Knapp and Michaels depend on: "You will either be ascribing these marks to some agent capable of intentions (the living sea, the haunting Wordsworth, etc.) or you will count them as nonintentional effects of mechanical processes (erosion, percolation, etc.)." To the contrary, a mad lib is the product of agents capable of intentions - me - but not the right sort to count as textual meanings. In the present case the meanings are unintended effects of intended mechanical processes. (You could say the effects are intended de dicto, not de re, if you like saying things like that.)
This leads on to another incidental point that may be nagging, because it is really a huge point. Back to Fish's op-ed:
The moment you decide that nature caused the effect, you will have lost all interest in interpreting the formation, because you no longer believe that it has been produced intentionally, and therefore you no longer believe that it's a word, a bearer of meaning.
First, the news that natural rock formations cannot be interesting to interpret will come as a crushing disappointment to geologist readers of the NYT everywhere. Second, my mad lib is very like the message on the rock; a random production. But it was produced intentionally. It's just that the intentions can't be its meaning. So Fish, too, is operating with a false dichotomy. Third, it is surely not the case that there is never any interest in interpreting even randomly-generated linguistic products. Whimsy is not a harsh mistress, but she can be a steady one. Fourth, in the mad lib case, the fact that this product has been made a stake in a philosophical debate - the fact that there are people planning and meaning things on both sides - lends it significance. It is a move in a game. We wonder about the implications. Ergo, authorial intentions make it meaningful, if you like. (Perfectly standard usage.) But these are not the right sorts of intentions, for Knapp and Michaels' purpose; see above. Furthermore, these are not the right sorts of meanings.
Let me distinguish four senses of 'meaning'.
1. Sentence meaning
2. Speaker meaning
What Knapp and Michaels are concerned to do in "Against Theory" is collapse 1 into 2, in a very strong sense. My mad lib is a counter example: a case of 1 but no 2. So 1 must be distinct from 2. But before resuming the push, it is noteworthy that Fish (and I think Knapp and Michaels as well) are neglectful of 3 and 4, as distinct from 1 and 2. This is shown by the oddity of Fish's failure to consider that a natural formation - particularly an odd one - might cry out for interpretation. We want to know what causes things, and what things imply about the world. None of this has anything essential to do with language or intentions to mean, but it is rather easy to get the two crossed. In my chess piece, I quote Pale Fire:
Reading from left to right in winter’s code:
A dot, an arrow pointing back; repeat:
Dot, arrow pointing back … A pheasant’s feet!
The fact that pheasant tracks imply pheasants is not a semantic fact. But, especially since we use 'imply' in Gricean sorts of cases - by x, someone means y; and because we are willing to interchange 'meaning' for 'implication', i.e. 'this means pheasants'; and because we use 'significance' when the implications turn out to be interesting ... well, the ground is laid for a failure to distinguish 3 and 4 from 1 and 2. In short, it is a bit unfortunate that the deep ethical question of the meaning of life is a homonym of the lexicographer's puzzle as to the meaning of 'life'.
Take the chess example (for longer treatment, see my chess piece.) As mentioned above, we can always ask what moves mean. This is shorthand for a twofold inquiry into plans and implications. We want to know what the intent of the mover is; we want to explore the combinatorial possibilities. Some suitable vector of these is a move's significance, you might say. A move is significant if the plan is good or bad or intriguing, and/or if the combinatorial possibilities are decisive; if the move is a turning point. (Trivially, every move implies a vast range of possibilities; not all of these are significant.) Yet none of this is meaning in either sense 1 or 2. To make matters more confusing, chess does have meanings in senses 1 and 2. Each move is, if you like, a sentence. Moving your knight to square d5 is functionally equivalent to announcing 'knight to d5', which is a sentence, and which works through your opponent's recognition that, by announcing that, you intend it to be recognized you are making a move etc. etc. But chess meaning has relatively little to do with either 1 or 2; the action is all in 3 and 4. To understand chess notation is to understand everything there is to know about chess meaning, in senses 1 and 2; but that is tantamount to understanding almost nothing about the meaning of a game, or move, or position, or opening, or combination, or plan.
Neither 3 nor 4 is a function of 1 and/or 2 - and not just in chess; in poetry and novels and life. So how can Knapp and Michaels and Fish hope to settle, decisively, questions of 'meaning' and interpretation, simply by handling 1 and 2? Possibly Fish would say he simply didn't have the space in his op-ed to deal with every such complication. He is simply concerned to collapse 1 into 2. This isn't good enough, but at this point there's a fork in the argumentative road. First, there are serious philosophical problems with collapsing 1 into 2, never mind about 3 and 4. The most philosophically interesting aspect of these problems may be confusion about the word 'theory'. Fish, Knapp and Michaels' usage is telling in a way highly relevant to our late Theory's Empire discussions. But an equally decisive way to run the refutation is just to show that practical criticism doesn't work the way Fish, Knapp and Michaels suggest. This can be done without too much heavy-duty philosophy. I handle that practical angle in part II. Go straight there if you're sick to death of philosophy and just want it over with.
Let's consider some standard philosophical objections to Knapp and Michaels' attempt to collapse 1 into 2. In "Literary Theory and Its Discontents", Searle makes the simple point - made by Wilson before him - that it is perfectly fine to regard something as a word in virtue of its shape. He quotes some concessions in this regard from Knapp and Michaels' 1992 response to Wilson and simply notes:
But once they concede ... that "the physical features of a set of marks intrinsically determine whether that set of marks is a token of a sentence type in a given language," then they have already conceded what they claim to be denying, namely that there are at least two types of meaning, the conventional sentence meaning and the intentional speaker's meaning ... This allows for precisely the possibility that they have been claiming to deny, namely, that a text can be regarded as either a string of sentence tokens and its meaning examined independently of any authorial intent, or a text can be regarded as a product of an intentional speech act and its meaning examined in terms of the intentions of the author. (p. 161)
This really ought to be game, set and match. But in The Shape of the Signifier Michaels attempts a response [you can search inside at Amazon and see context]:
As Searle eventually put it, "In linguistics, philosophy and logic words ... are standardly defined purely formally"; hence, "it is simply not true that in order for a physical token to be a word ... it must have been produced by an intentional human action." The appeal to t he "purely formal" here is an appeal to the physical, to the shape of the signifier. Clearly the marks are shaped like language - they look just like English words. But is being shaped like language enough to be language? Is what makes a word the fact that it's being used as a word or the fact that it's shaped like a word? (p. 58)
Michaels considers a case in which we find giant stone letters on the surface of Mars, suspiciously Wordsworthian when regarded from an orthographico-low ortibal point of view. (Michaels is in the process of discussing Kim Stanley Robinson, so there is a sort of motivating fictional occasion for the Martian landscape, but a rather strained one.) But now suppose that there is some incidental problem recognizing whether one shape is more like an 'r' or an 'a'. This is supposed to be a problem:
For if shape were decisive ... then something that looked like an r would necessarily be an r.
But what are the criteria for looking like an r? There aren't any, not because we don't have some idea of what an r looks like but because, even if something looks to us like an a, we don't have any argument against someone who says that the same thing looks to him like an r. How could we? (p. 59)
After some more notes in this vein:
Are there letters on Mars? It depends on your perspective. But which perspective is the right one? You might be able to answer the question which perspective is the most beautiful or which perspective is the most interesting, but you obviously can't answer the question which is right ... The question of whether those formations really are letters regardless of your perspective makes no sense, since as long as the relevant criterion is formal (its shape), the question of whether the formations really are letters is a question that is crucially about your perspective ... Because what something looks like must be what it looks like to someone, the appeal to the shape of the signifier is at the same time an appeal to the position and hence to the identity of its interpreter. (p. 60)
The first thing to say in response is that Searle already went over this: "From the fact that every syntactical token is a physical entity, such as an acoustical blast or a physical mark, it does not follow nor is it the case that syntactical categories are categories of physics. Notions such as "sentence of English" cannot be defined in terms of, for example, acoustics or mechanics" (p. 163). Searle says it is because something's being a sentence only makes sense against what he calls the Background. Mostly, this is the point about occasionalism made above. The fact that someone had to make a language by intending something does not imply that intention is needed for each linguistic occasion. But there is more, and I don't think Searle really puts it as clearly as he could.
The thing Knapp and Michaels crucially miss, from the start, is that word and sentence and so forth are vague categories. These terms are family resemblance terms, as Wittgenstein would say. This means there will be clear yes and no cases and some grey area cases. (It also means linguists and philosophers and other interested folks can perfectly well carve out sharp definitions of wordhood and sentencehood, from the more or less cloudy mass of ordinary usage. The warrant for such sharpened concepts will be utility relative to some intellectual or systematic project. Knapp and Michaels' suggestion that their argument effectively ends the project of trying to come up with just such theoriers precludes them saying this is what they are doing, however.)
The criteria that determine membership in the family of sentences and the family of words and so forth fall fairly neatly (but probably not completely neatly) into two piles: intentionalistic and formalistic. A paradigm case of a word or sentence is one in which something of a standard shape is backed up by a standard sort of intentionality. A paradigm case of a non-word or non-sentence is something of a non-standard shape, without appropriate intentionality. Borderline cases are those in which there is something peculiar and non-standard about one or the other or both.
Some examples (you could do this yourself, but I am seeding a few further points):
Above I mentioned that a move in chess can be regarded as a sentence. Reason: it is a symbolic act, and precisely functionally equivalent to a specifiable sentence that encodes specific and highly determinate propositional content. But we are reluctant to say a move is a sentence because it doesn't really look like one. It is highly non-standard, formally. So is a move in chess a sentence or not? I can't see that there is any reason to go either way, unless someone has a good reason to stipulate in or out. (Linguists might prefer not to have to deal with such cases, hence might find a way of ruling out chess moves as sentences. That would be fine.)
Suppose I am flying high above a river and I look down and say: it looks like God himself signed his name to the land; that is, if God were a doctor signing a prescription slip - just a squiggle, nothing remotely resembling actual letters. (Get it? There's a reason Wittgenstein said a book of philosophy could consist only of jokes. The joke hinges on the vagueness of criteria for wordhood. Hey, he didn't say funny jokes.)
The Nabokov case. If someone wants to pretend that there is a
'language' of animal tracks, it helps that they tend to appear in lines
consisting of discrete, distinctively shaped scratches. It's a kind of
natural pun on the sorts of things humans regard as writing. It looks
like a word, and it 'means' something. We will want to say it isn't
really a word, but if someone started to insist, the usage might take
root. If we found a language in which the word for (what we call) words and the word for spoor were one and the same, we would understand.
Suppose I write out sentences in English, lightly encoded, so A = B; B = C; C = D ... Z= A. 'The cat is on the mat' comes out: 'Uif dbu jt uif mbv'. Now is that written in English? Well, yes. Probably. Is it an English sentence: yes and no. Is 'uif' a word in English? One wouldn't say so. (Trying this in Scrabble will get you nowhere. But really any of these questions could be answered either way you please, if you have a special reason. The reason it is no-go in Scrabble is that it would obviously wreck the game.)
Obviously, formal definitions of words in terms of shape are complicated by the existence of strange fonts, and the possibility of greating indefinitely more strange fonts than the world has already got. You simply couldn't formally define a range of possibly acceptable fonts. You need the fact that they are intended to unify the set.
Now, epistemological twists. Consider the resemblance questions Michaels raises. Suppose I have a stick with a sharp bend in the middle. The bend is not sharp enough for it to look like a V, but if I put it in water, just right, it will look like a V to a human observer. So does it look like a V or doesn't it? Well, it seems to me there is no real problem with saying it doesn't, until such a time as we start doing a lot of semi-submerged writing; that is, until a very strong infusion of intentionality tips the weight of family resemblance the other way. Plus, if we got used to dealing with shapes intended to look like V's in water, those shapes would start to look to us like V's even out of water, because we would be attuned to their purpose.
Suppose there is a stone formation that looks like Richard Nixon. Or a face on Mars, instead of a lot of Wordsworth. In general, take any manifestation of pareidolia. Is there any reason to think that there is a problem saying these things really resemble what they seem to resemble? Notice I am returning to Knapp and Michaels' odd doubling of levels of resemblance in their original formulation of their conclusion: "They will merely seem to resemble words." Why not say they really resemble words? Well, we say seems to resemble because we want to cue our audience of our awareness that in these cases there is a distinctive cognitive contribution of our species. But many of our concepts may be - to pick another Martian example - grokking concepts, as Steven Yablo calls them (PDF). (See this Brian Weatherson post for an introductory discussion of the notion. Brian is skeptical. It all makes sense to me.) There doesn't seem to be any need to signal in every case that we are applying a concept in a way only a human being would; presumably our audience knows we are human. (Again, you can tell stories where this isn't the case. But then our usage may change.)
These are all cases in which there is something odd about form - the shape of a signifier, if you like.
I'm looking for a Wittgenstein quote that pretty much says it all: 'One of the main criteria of something's being a word is: looking like a word.' Something like that. (Anyone have the reference?) Not the sole criterion; one of the criteria.
What about intentionality? Well, the mad lib case and Knapp and Michaels' and Fish's cases are borderline cases. I leave it to the interested reader to multiply them - all day long, if you like.
What is the moral of the story? It is twofold. First, the meaning of words like 'word' and 'sentence' is vague, in ordinary usage. Clear cases, borderline cases. Knapp and Michaels' original argument depends crucially on this not being the case. If the question of whether the scratches in the sand are words waits on an uncertain weighing of the relative importance of formal and intentional factors, they are refuted. This problem is compounded by their attempt to address Searle by treating his position as cartoon formalism:
If we need to know what language the author wrote in, we need to know what the author intended, and once we care about what the author intended the rules of the language are relevant only because they help us to figure out that intention. If, on the other hand, our appeal to the rules of the language is not a disguised intentionalism, if we are, like John Searle, more purely formalist - that is, indifferent to the question (by whom or by what?) of how the marks or noises we interpret were produced - then we treat them perforce as accidents and, once again, do not interpret them at all. This is the point of de Man's "radical" formalism: he sees that the only alternative to thinking of "Marion" as meaning whatever (more or less complicated thing) Rousseau meant by it is to think of it as meaning nothing - as a rock instead of a speech act. (p. 117-8)
Some of the mistakes here have been dealt with already. The question of whether you wish to interpret something does not depend on whether you see intention behind it; it depends on whether you find it interesting. The de Manian opposition is clearly false. You needn't think of my mad lib as meaning nothing, although it certainly lacks an author. Finally, Searle is patently not a 'pure formalist'. He sees the question of whether something is a word, or sentence, as one that requires us to consider both intentional and formal factors. The fact that he is willing to invoke the necessity of the Background gets him off any hook concerning pareidolitic effects and such.
Knapp and Michaels simply never consider the obvious possibility of mixing criteria - intentionalist and formalist; let alone that there might be vague concepts. There is something oddly Derridean about this studied avoidance of (what seems to me) the only plausible position. But that is a subject for another post. [UPDATE: Matthew Yglesias agrees this is a surprisingly common problem.]
These reflections lead us on to the second line of defense Knapp and Michaels' have tried out. I will simply quote Searle's encapsulation and refutation of it, for starters:
Nor will it evade these inconsistencies to say that Knapp and Michaels are interested in the problems of literary theory only as an "attempt to govern interpretations of particular texts by appealing to an account of interpretation in general," and that they were not interested in these abstruse questions about language in general. This will not do for two reasons: First, they do in fact make claims about language in general and not just about literary texts. Indeed, they even criticize other authors - me, for example - who are not especially concerned with literary texts, and they use examples such as "My car ran out of gas" that have no special connection with literary texts. (p. 162)
This is Searle's response to a criticism of his paper that is raised by Kieran Setiya in Sean McCann's comment thread. Kieran, by the by, is a philosophy professor with a blog. I just discovered it. And he blogs about valvish sort of things. For example this highly amusing review of Eagleton's After Theory. (Into the blogroll with him.)
But he's still wrong about Searle. Setiya points out, basically, that Searle grants intentionalism as a literary critical strategy.
If Searle is right that the only alternative is speaker meaning, that this is a matter of the speaker/writer’s intention, and that this identity is “trivial”, then his conclusion is exactly the same as Knapp and Michaels’. All he’s done is to iron out a wrinkle in their argument: their carelessness about sentence and speaker meaning. Searle’s essay contains no “devastating blow” to their intentionalism about literary interpretation, and in fact defends it.
This is half right. I think Searle does end up in the same intentionalistic place (notwithstanding Michaels' late tendency to brand him an extreme formalist); but Searle also says that there are two different ways one might understand "text" - formally or an an intentional product. So he acknowledges that one could read while ignoring intention, however odd that would be, whereas Knapp and Michaels deny the possibility. (The difference between saying something would be odd and saying it would be impossible is rather large.) But more than that, he makes the point that Knapp and Michaels' stake their claim not on literary critical grounds but on general philosophy of language grounds.
So their point about the indistinguishability of sentence and speaker meaning is not some wrinkle. It is absolutely the whole cloth. Take that away, and there is quite literally nothing left to their view that couldn't be printed on a fortune cookie fortune. Just: intentionalism seems plausible. (I suspect Setiya does not know this about Knapp and Michaels - no particular reason he should.)
This is a good point to note the only really serious unclarity in Knapp and Michaels' paper - namely, their usage of the word 'theory'. They are actually quite clear in declaring what they mean. But they might have helped their readers more by noting how their usage inevitably generates blurry double-vision, when crossed with ordinary usage, which has to be present as well. The first line of their paper: "By "theory" we mean a special project in literary criticism: the attempt to govern interpretations of particular texts by appealing to an account of interpretation in general." This explains why their theoretical argument (what else to call it?) that textual meaning must be identical to author's meaning isn't a 'theory', in their sense: because it does not entail any practical, normative consequences. Their theory does not imply that anyone is right, or anything is wrong. So it is not a theory, in their sense.
An analogy: suppose someone said, "by biological 'theory' we mean any general account of biological processes that has practical effects on medical science, i.e. that implies that some doctors are right and some are wrong, or that pharmaceutical companies could be making better drugs, etc." Or suppose someone said, "by physical 'theory' we mean any general account of the material universe that helps us build better technology." These would be extremely odd usages because of how they would inevitably cross-cut the ordinary use of 'theory' to denote general explanatory accounts (spell it out how you like.) Ask a biologist whether he is doing biological theory and he may have to say: I don't know, because I don't know whether my general explanatory account of such-and-such life processes will imply norms for medical treatment.
Turning the point around, it is clear that an attempt to govern interpretation by an appeal to an account of interpretation in general needn't be at all theoretical, in anything like the ordinary sense. Suppose I say, 'you should accept my interpretation of "A slumber did my spirit seal," because I am always right about what poems mean.' Expressing this obstinate dogmatism would count as 'doing theory' because it attempts to govern interpretation with respect to a general account of interpretation. (Saying I am always right is quite general, you see.)
Getting back to the point that Knapp and Michaels can hardly retreat to the position that they are doing something narrowly literary critical, i.e. theoretical (in their sense), whereas Searle is dealing with philosophy of language in general: no, because Knapp and Michaels are "against theory" in their sense. They are doing it in Searle's sense. Searle's point is that they are doing it badly. Any philosophy of language that botches the elementary distinction between sentence and speaker meaning is bad.
This point would not be worth harping on if it did not have rather general consequences for 'theory' - as in The Norton Anthology of Theory and Criticism. There is something peculiar about the fact that Knapp and Michaels' "Against Theory" is included but no work by Searle or any other analytic philosopher of language (except Austin.) You might say it is due to the fact that analytic philosophers are pitifully unsophisticated hence unworthy of inclusion. I don't think this is really plausible. Anyway, the 'unsuitable style' argument does not fly because Knapp and Michaels' style is not distinguishable from that of someone like Searle.They offer general conceptual arguments, thought-experiments, etc. It is surely irrational to include them just because they are English professors (why should English professors automatically be better at philosophy than philosophers? Shouldn't there be at least an argument to this conclusion?) Anyway, there are lots of old philosophers in the Norton, just few contemporary ones from outside the continental tradition. Even though analytic philosophy has a massive and quite venerable tradition of theorizing about meaning and intention and so forth. Why are they ignored? (I don't mean in a sociological sense. I mean: what is the actual intellectual justification for the selection?)
The answer is that Knapp and Michaels are included because they are regarded as 'doing theory' in the sense that they officially declare impossible (complex irony). They are part of a tradition of having effects of a certain sort. (In this limiting case, the effect of declaring effect impossible; which may have a big effect.) Analytic philosophers may have vastly better and more rigorous and sophisticated theories of meaning and intention - no showing has been made to the contrary - but they haven't really wowed the literary critics. So they haven't had effects. So they don't 'do theory'. So they don't get in the Norton. A quote from Jonathan Culler, Theory and Criticism After Structuralism. This was written in 1982 and I quote at length because it really says a lot about 'theory' in the sense that we've been discussing it for two weeks:
Recently ... there has been increasing evidence that literary theory should be conceived differently [than as the elimination of methodological error, so as to facilitate valid interpretation]. Whatever their effects on interpretation, works of literary theory are closely and vitally related to other writings within a domain as yet unnamed but often called "theory" for short. This domain is not "literary theory," since many of its most interesting works do not explicitly address literature. It is not "philosophy" in the current sense of the term, since it includes Saussure, Marx, Freud, Erving Goffman, and Jacques Lacan, as well as Hegel, Nietzsche, and Hans-Georg Gadamer. It might be called "textual theory," if text is understood as "whatever is articulated by language," but the most convenient designation is simply the nickname "theory." The writings to which this term alludes do not find their justification in the improvement of interpretations, and they are a puzzling mixture. "Begininning in the days of Goethe and Macaulay and Carlyle and Emerson," writes Richard Rorty, "a kind of writing has developed which is neither the evaluation of the relative merits of literary productions, nor intellectual history, nor moral philosophy, nor epistemology, nor social prophecy, but all of these mingled together in a new genre".
This new genre is certainly heterogeneous. Its individual works are tied to other distinctive activities and discourses: Gadamer to a particular strand of German philosophy, Goffman to empirical sociological research, Lacan to the practice of psychoanalysis. "Theory" is a genre because of the way its works function. The practitioners of particular disciplines complain that works claimed by the genre are studied outside the proper disciplinary matrix: students of theory read Freud without enquiring whether later psychological research may have disputed his forumulations; they read Derrida without having mastered the philosophical tradition; they read Marx without studying alternative descriptions of political and economic situations. As instances of the genre of "theory", these works exceed the disciplinary framework within which they would normally be evaluated and which would help to identify their solid contributions to knowledge. To put it another way, what distinguishes the members of this genre is their ability to function not as demonstrations within the parameters of a discipline but as redescriptions that challenge disciplinary boundaries. The works we allude to as "theory" are those that have had the power to make strange the familiar and to make readers conceive of their own thinking, behavior, and institutions in new ways. Though they may rely on familiar techniques of demonstration and argument, their force comes - and this is what places them in the genre I am identifying - not from the accepted procedures of a particular discipline but from the persuasive novelty of their redescriptions. (pp. 8-9)
This is clearly the sense of 'theory' that is operative not just in Knapp and Michaels but in the Norton. (Note the connection with Michael Bérubé's recent post: it's all about the ostranenie.) There are all sorts of puzzles and peculiarities hereabouts. For example, 'persuasive novelty' has nothing essential to do with validity, soundness, truth, correctness, or intellectual quality; not necessarily; so there is no reason why good 'theory' should be intellectually good. But, of course, Knapp and Michaels, in arguing that 'theory is impossible', are in effect importing standards of intellectual quality. They clearly think it's important that their argument be a valid argument, in the philosophy of language and mind. And Culler would surely want to insist that intellectual quality is important. Theory isn't just saying strange things, however stupid, that bowl over the rubes.
Yet it's not clear how Culler can possibly build the quality requirement in, given his characterization. He writes a bit further on: "Literary theorists may be particularly receptive to new theoretical developments in other fields because they lack the particular disciplinary commitments of workers in those fields" (p. 11). Now what can this possibly say? I think it must say that literary theorists are more easily bowled over by strangeness, and more likely to see the possibility of bowling over others. Their lack of disciplinary commitments can only mean they are less fastidious about intellectual quality control, in a standard 'truth and validity and evidence' sort of way. What else can he be saying? But why should that be a good thing? Culler won't want this, but it seems like a recipe for offering strange, bad arguments.
Which gets me back to Knapp and Michaels.
My points are starting to spread out. Let me try to focus. Consider again the case of biological 'theory' as everything that might have an effect on medical treatment. (Let's even grant that we are only concerned with good effects, although no one has quite made clear why we should only be concerned with those, given the definitions we are dealing with.) You can't possibly teach this 'theory' as a foundation for medical treatment. I don't see how you could teach it neat. Because to understand what you are teaching you would need to understand theory in the ordinary sense. That is, you would have to study biology. You need some general explanatory schemes. This is what makes 'theory' - in Knapp and Michaels' and Culler's sense - so peculiar. You need not have a high opinion of Searle and co., but you still can't possibly be interested in Knapp and Michaels essay without having an interest in a vast body of literature that, apparently, isn't being considered: philosophy of language. So it seems to me that including something like Knapp and Michaels' "Against Theory" in your anthology is not a stable position. Either you really aren't interested in philosophy of language, in which case why should you care about their argument concerning philosophy of language? Or you are interested in philosophy of language, in which case why aren't you studying philosophy of language much more extensively?
In the final version of my Nabokov piece I make more or less these points with respect to Fish and Hirsch and some others. I point out that there is a quite general tendency in a lot of literary critical theory to try to get a decisive angle on interpretative questions by proposing 'to answer a few general questions in philosophy of language', or something of the sort. Usually a lot of heavy-duty epistemological 'problematizing' is brought to bear. One of the papers in The Literary Wittgenstein - very good paper, by Martin Stone - puts this well with regard to Stanley Fish.
Fish says that he aspires to a "severe [theoretical] minimalism," and that "this parsimony of ambition distinguishes [his] from almost any other argument in theory." But "so much for parsimony," I feel inclined to say, when Fish nominates "interpretation" as a general condition of the possibility of a text's determinately meaning one thing rather than another. Such "interpretivism" (as I shall call it) looks like nothing less than metaphysics in the classic sense: an attempt to lay bare the conditions of intelligibility of the world as a whole, of everything. The "implications [of the ubiquitous need for interpretation] are almost boundless" - Fish says - "for they extend to the very underpinnings of the universe." Are we really supposed to regard this as a bit of hard-won pragmatism, fashioned to combat other suspiciously metaphysical pictures of meaning? ... In Fish's argument ... "interpretation" begins to look like another name for - an occupant of the same explanatory place as - divinity: it is the terminus of all other explanations of meaning, the condition on which they depend. (p. 189, brackets in the original)
My choice of a theological doctrine - occasionalism - as a suitable analog for Fish's and Knapp and Michael's view was not accidental, you see. What Stone is marveling at is that Fish can call himself a theoretic minimalist while engaging in truly hair-raising theology-grade metaphysical speculations; which is certainly the case. But the answer is that Fish is using 'theory' in Knapp and Michaels' sense. His divinity is very laissez faire, not making normative demands. But interpretation is still, for Fish a Theory of Everything, even if this Theory of Everything is not a theory at all. Ah, puns.
The problem here, quite apart from needless confusion generated by odd usage, is that the Theory of Everything on offer is unsound philosophically (whatever its effects). Fish (and Knapp and Michaels) are too preoccupied with literary cases. They suffer from a one-sided diet of examples. In my Nabokov/chess piece I focus on chess notation - simple algebra. Here are sentences; the meaning of the sentences is completely determinate without any need to consult authorial intention. It may not be that English works the way chess notation does, or legal cases. But now we are arguing cases. Take Fish's complaints about theories of legal interpretation. He is attempting to prove that Scalia is wrong not by practicial considerations about the law but by perfectly general arguments about philosophy of language. Fish and Knapp and Michaels deny the possibility that any language can possibly work the way chess notation plainly actually works. Actual implies possible. Game over.
The other point of picking chess notation is that a chess game is a beautiful, highly interpretable linguistic objection; in short, a poem. This is the burden of my dialogue. It is tempting to assume that poetic effects must be functions of linguistic complexity, or of intentions, but this clearly doesn't work for chess notation. It is hard to run down the list of formalist candidates, but it seems sufficient to note that chess notation has no rhetorical function. It does not aim to persuade, nor does it admit the possibility of figuration. (You can say it has the rhetorical function of clarity, if you think that removing misunderstanding is a rhetorical function, as Wayne Booth sometimes says. Chess notation is perspicuous and easy to read and learn. But this is not going to explain the interpretability of a chess game.) And obviously intention isn't the answer. The meaning of chess is not a function of what any player intends, because players can miss things.
But aren't there all sorts of metaphysical and epistemological problems about how chess is even possible? Yes. But it hardly seems necessary to get into all that, now is it? We are just doing literary criticism here.Continue reading Near Science II (Bits and Pieces) ...
Interesting post. I agree that the ambiguity of terms such as ´intention´, ´intentionality´ and ´meaning´is an important part of the problem with the ´intentionalist controversy´ (Gibbs´ term).
Notwithstanding the interesting reflection on the ´form/convention/intention´-dispute, I was especially struck by the final part of your piece. The lack of dialogue between analytic and continental philosophy (of language) is indeed striking and detrimental to research (both in philosophy and literature departments). As I have pointed out in an earlier comment, there is some interesting work being done in the relatively new field of ´philosophy of literature´ (basically a reorientation of ´aesthetics´). As of yet, it does not seem to have made much of an impact in English departments, however. There seems to be a similar (and similarly regrettable) lack of dialogue, moreover, between literature and linguistics departments. Although I took some introductory courses in linguistics, I have always felt that I did not know nearly enough about Saussure, Chomsky, Searle, Grice and cognitive linguistics to properly evaluate certain claims in literary theory.
As regards the K & M-piece, I agree that its stylistic clarity tends to undo the theoretical vagueness that we are all interested in. As you suggest, however, this clarity has the advantage of allowing us to spot the essay´s theoretical weaknesses. Thinking about your suggestion that Theory is a sort of inter- (or, perhaps better, extra-) disciplinary discipline that does not have any clearly delineated standards for evaluation, I was thinking that some works of Theory are equally weak, but use an obtruse style to get away with it. Although that might be down to my ignorance, that was the impression I got when reading Lacan´s piece on the Mirror Stage. This is not to say that I do not appreciate stylistically complicated theoretical works or that I want to champion empirical research, however. I love certain works of Theory, but I nonetheless hate certain others (cf. the earlier Valve-post) and I think that this is in part down to these stylistic issues (cf. my unease about the Lacan-piece; it seems, well, bad, but I should probably read up on his work before being able to judge this single article properly, which is hardly an ideal situation). Maybe the question, therefore, is how we can properly evaluate works of Theory.
Fine, I admit it: I cannot see your point. I think that Kieran Setiya is right – which means that Knapp and Michaels (and Fish) are right. They may be trivially right, but they are right nonetheless.
It doesn’t seem to matter at all that certain collections of marks can look like English words or sentences and that these sentences can have what one might call a “standard meaning.” But this just seems to say that when we usually see collections of marks that look like X, they usually mean Y. That’s the function of any dictionary. Such predictability is what may language work as a communicative tool. English speakers tend to use words like “cat” and “dog” and “theory” (as opposed to “fub” and “ugiou”). And when they use words like “cat” and “dog,” those words tend to have a semi-stable collection of meanings. Knapp and Michaels have said as much.
But we are talking here about – and only about – the meanings of particular utterances. Dictionaries and resemblances and the predictable patterns of English usage can (often) help us to interpret such utterances. But the fact remains that when we try interpret what utterance X means, we are always reduced to making educated guess about what it’s author meant by that utterance.
Allow me to run through some of your examples, and add a couple of my own:
*** You can treat pheasant tracks “as if” they were language – but the moment you do so, you are treating they “as if” there were intention informing those tracks. To me, cuneiform script looks a lot like pheasant tracks. But I treat the former as language the moment I decide that these marks in the clay tablet were created intentionally. And now, my job is to decide what the meanings of those utterances are.
*** The sentence “Knight to d5” also has a standard meaning – indeed, a bunch of standard meanings, which vary from situation to situation in (fairly) predictable ways. But the crucial issue is what a particular utterance or inscription of “Knight to d5” means. Am I announcing my next move, telling you or the computer to make that move? Or am I merely muttering possibilities to myself under my breath? Or trying to distract you? Or asking for a soda? How will you decide what this particular utterance means? Right: you will try to figure out – by asking me, by making a best guess, by consulting a dictionary – what my intentions are.
*** My son used to pound on my computer keyboard when he was a baby. Occasionally, English words (and other languages too, I bet) would appear on the computer screen. One time, he even typed the letters “I AM” – admittedly with a bunch of gibberish on either side. I know what this set of marks usually means. But what does this utterance mean? Is it meaningful? How could it be meaningful without intention?
*** Think of the Bible Code dust-up from a few years back. People were finding words and phrases “hidden” in the text of the Bible by looking at regularly spaced characters. Let’s say I try it out myself and find that that if I take every 2000th letter of the King James Bible, it spells out “Peter will die today.” Sure, it’s an English sentence. And when sentences like these are usually produced, they usually mean such and such. But what does this utterance mean. Put simply: people were only interested in the Bible Code when they thought there were intentions behind these messages. If they were just random collections of letters, no one would have cared – even if those collections did create “real” English words.
*** I goof up “left “ and “right” all the time – so I’ve taken to pointing when I give directions. This means that, frequently, when I tell you to “turn right at the light,” I really am telling you to turn left. I want you to turn left. I intend to convey and communicate that command. Something goofy just happens. I end up saying words that almost always have the meaning “turn right.” Nonetheless, my particular utterance of that sentence means “turn left.” It’s not a great way to run a language – violating all the predictable rules of English usage. But it is what that utterance means. And people who have driven with me a long time know this. When I say “turn right” and point to the left, they know that I mean “turn left.” Or they know enough to ask for clarification about the (intended) meaning of that utterance.
*** Finally, my son now makes up new words all the time – words that are not English and are not in any dictionary. His latest one is “Kentuckish.” The word seems to mean “of or related to the inhabitants of Kentucky and their unique features.” Use it in a sentence: “That’s funny. He doesn’t look at all Kentuckish.” (Sure, it looks and sounds like “Kentucky.” But for the sake of argument, he could have invented the word “Blippish.”) That utterance has a meaning. That meaning resides in my son’s intention.
I admit, I have all but ignored your focus on “implication” and “significance.” I just don’t see what those aspects of meaning have to do with the question at hand – the meanings of particular utterances. Sure, we can say that they potato that looks like my grandfather has implications (for me, about soil conditions, etc). It may even have significance, because it reminds me of my grandfather. But I will not say that that the potato has “meaning” in the same way that “Kentuckish” has meaning.
It may be a simple, trivial thing. If we all agree on that trivial aspect of language, then fine. We can stop right here. But that trivial truth still seems to pack a punch and to evade many readers. Authorial intention may be a simple rule to the meaning game, but it’s still seems to me to be the only game in town.
And as a case in point, take your own reaction to my comments. How will you decide what they mean (as opposed to their significance)? Or if I have drastically misunderstood your original post’s meaning, what exactly have I misunderstood? If I “cannot see your point,” what exactly have I not been able to see?
Sorry for the cross-posting with “Near Theory II,” but I thought my follow-up comment would make more sense here, on the tail of my first note
After posting my comment above, I realized how little attention I had paid to your “Mad Libs” example. I did so, in part, because I thought that the set-up did not really affect the pay-off in interesting ways.
I do think that the beginning scenario does have interesting implications for the topic of authorial intention because of how it “splits up” intention and the creative act – somewhat like an exquisite corpse.
Earlier respondents to Knapp and Michaels raised similar points, asking (for instance) about the “author” of what seem to be group efforts, such as the staging of plays. Is the performance the utterance? If so, how do the separable intentions of the playwright, director, and actors affect the meaning of that utterance?
But by the end of your post, you are trying to interpret the meaning of the “Mad Lib” poem, the “blanks” of which may well have been filled in my a computer. And you certainly seem to be able to find meaning – to create interpretations that “unify” the poem and find connecting “themes.”
But so what? You say that you add in just a pinch of “as if” authorship. (What might an author have meant had he or she uttered such-and-such a poem?) But, in my eyes, that pinch is enough. Without that pinch, there’d be nothing worth doing. And to the extent that this is a randomly generated poem, the act of interpreting it really does seem pointless, doesn’t it? The whole thing is an “as if” game. (Consider, once again, the Bible Code predictions I find hidden in the New Testament or in Moby-Dick. Is it possible to find patterns and potential significance? Sure. But why would I do it unless I thought the utterance were authorially meaningful—unless I thought that someone was trying to tell me something?)
In fact, I wonder if Searle’s own earlier work presents another vision of the same point. Think of the Chinese Room Experiment. The man inside the box clearly is able to produce sentences that look like Chinese – that can be read by literate Chinese speakers, that can be looked up in Chinese dictionaries, that can be translated into other languages, etc. None of that is under contention.
But do we still want to say that the Chinese utterances made by the man in the box have the meanings that the people outside the box are ascribing to them? The boxed man writes squiggle-squiggle-squoggle –- marks that usually mean, “I understand Chinese” or “Help, I’m trapped in a box” or “I want a pizza.” But do we want to say that these particular utterances are meaningful in this “standard” (or even non-standard) ways?
After all, taking the first example, the boxed man does NOT understand Chinese. Neither did he intend to lie about that fact. In fact, he didn’t intend to communicate any meaning at all when he made those squiggly marks. If the people outside the box interpret those marks to mean that the man inside the box understands Chinese (see utterance one, above), then they are mistaken. Understandably mistaken—but mistaken, nonetheless.
(The question of “Is it Chinese or does it only resemble Chinese?” seems a small side issue, even if some in this debate would disagree. My son and I can create meaningful sentences that contain words that exist in no “language.” See Davidson on this point.)
As Searle argues, physical marks and even syntax is separable from semantics. And semantics are all that matters in this instance. I believe that Knapp and Michaels and Fish would concur.
But perhaps I have misunderstood them.
Peter, I may misunderstand, but I think the answer to your challenge appears in John’s reference to the fact that Searle “acknowledges that one could read while ignoring intention, however odd that would be, whereas Knapp and Michaels deny the possibility.”
There’s a difference btw describing an activity as odd, or silly, or unworthy, and declaring it impossible. In fact, though this wouldn’t be my view (since my inclinations run the same way as yours, I think), the case might be made (Ray or Matthew or Luther Blissett might make it on this site) that there’s actually something valuable about engaging in the kind of speculative critical operation John performs in Part II--because it gives aesthetic pleasure, or cognitive exercise, or even ethical training.
You’re right, Sean. There may be less disagreement here than meets the eye. I suppose I would say (with you?) that “reading while ignoring intention” would boil down to one of two interrelated possibility.
Either you are deciding what certain sets of words usually mean when uttered by speakers of Language X. Or you are playing out an elaborate “as if” game –- spinning through what a non-existent speaker might have meant (i.e., practically anything) or what a set of words might have meant to you.
But these options seem to present either a fairly attenuated vision of “reading” on the one hand (since you are merely playing the linguistic odds) or a too-expansive version of reading on the other hand (since anything potentially “significant” would count).
But I’m not sure if even Michaels would say that such activities are “impossible.” They’re just different activities, valuable or not. If I am working out the particular meaning of some particular utterance, I am perforce driven to questions of intention.
"If I am working out the particular meaning of some particular utterance, I am perforce driven to questions of intention.”
True, but some critics hold -contra K & M- that the nature and retrieval of that intention can take two different forms, dependent on whether we are dealing with a spoken dialogue or a written (especially literary) text. Although I agree that every form of interpretation implies some sort of intention-attribution, I am not certain that every utterance implies the search for a single type of intention. Surely there is a difference between your interpretation of my post and your interpretation of ´Ulysses´ in terms of the author and goal you postulate me and Joyce having while you are reading my and his words (and, well, there is a difference in quality as well, of course)? That is not to say that you could not bring a literary interpretation to bear on this post, but simply that there are two ´genres´ of interpretation, so to speak, one with a clearly defined and one with a less clearly defined goal. It seems to me, in other words, that we look for the ´actual´ author´s intention in conversation, whereas we look for a -not any, of course- ´hypothetical´ author´s intention in reading (Levinson´s distinction, if I remember correctly).
In fact, though this wouldn’t be my view (since my inclinations run the same way as yours, I think), the case might be made (Ray or Matthew or Luther Blissett might make it on this site) that there’s actually something valuable about engaging in the kind of speculative critical operation John performs in Part II--because it gives aesthetic pleasure, or cognitive exercise, or even ethical training.
It is, in fact, loads of fun, especially when practiced with others, in real time. (The last time I did something like this, it was an interpretation of the logo on a guy’s shirt on a train. Fortunately he didn’t speak English.) In which case the positing of an author can be a strategic move, if you don’t like the way another person’s interpretation is going: such-and-such couldn’t be an allusion to Verdun, the author died in 1875.
You could call the Oulipians to your side, too. Queneau could not have consciously intended the meanings of all the possible sonnets in “Cent Mille Milliards de poemes” since there simply isn’t enough time in the world to construct them all. The S+7 algorithm for constructing texts (which is sort of like your mad-lib version of Wordsworth) depends crucially on which dictionary you use, which is unpredictable and therefore cannot be intended.
I think I understand the difference you describe, Trickster, but I’m not sure I feel its force.
On the one hand, it seems that you are presenting two different decisions that we make about intention, based on assumptions that we readers have about the author and the utterances in question. Do we think the author (Joyce or not) is intending to be clear and straightforward and predictable? Or do we feel that the author (Joyce or not) is intending to be allusive, elusive, ironic, figurative, punning, sarcastic, and so on? We can make these decisions based on our past experience with said author and the way he or she tends to say things. Or we can look for clues in the words. Or we can explore the genre in question and its literary history. But it’s still just a shot at intention.
Perhaps, to borrow liberally from Nelson Goodman, we sometimes decide that certain authors, genres, and/or utterances are more intentionally and semantically “dense” than others. Perhaps we decide that, with certain authors and utterances, every difference—in words, the number of syllables, the spelling, the sounds, the symbolic connotations—potentially makes a difference. (Joyce, for example, seemed to worry about every thing, every mark.) With other authors and utterances, we decide that the message is less dense and more “disjunct” or “attenuated”: every difference does not, we decide, make as much of a difference. Still, if we’re shooting at meaning, these are decisions about intention and what the author was trying to “do” with the marks she or he produced. (Even Joyce probably wrote shopping and to-do lists, as Gary Larsen has pointed out.)
One the other hand, you also talk about two different “goals” in interpretation –- one direct and simple, the other playful and open-ended. When we are talking to the IRS or our life partners, for example, we are usually worried about interpreting their meaning correctly. These things matter. Our livelihood rides on getting it “right.” But when we’re just talking about literature –- upon which very little money and happiness depends –- we have the freedom to just goof around, playing with words and ideas at hand, searching for something emotionally and intellectually satisfying. That certainly is a different goal, and maybe a different type of “reading.” I’m not certain.
I am not trying to make your ideas seem odd. Perhaps this is exactly what we do; perhaps we’re paid to play. But I’m not sure I’d give any of my students a good grade for doing it.
Sean, I just want to clarify my position on what you wrote above: “. . . the case might be made (Ray or Matthew or Luther Blissett might make it on this site) that there’s actually something valuable about engaging in the kind of speculative critical operation John performs in Part II--because it gives aesthetic pleasure, or cognitive exercise, or even ethical training.”
I don’t think of myself as a sort of Pater-ite, though I do love to read what might be called associative or impressionist criticism. One of the reasons my dissertation is not historicist is that I do think the main goal of writing about literature is to unpack and follow the meanderings of meanings in the text itself. I am interested largely in what a text says about the world and not in what a text makes me think of or feel.
At the same time, I do think that, as Bakhtin writes in “Discourse in the Novel,” an author sends words like beams of light through the air, and a critic can act as a sort of social prism, to separate that word into its various social resonances. Which is to say, a poem about freedom might be intentionally only a response to one position on freedom at that time, but the debate about freedom is the larger context in which the words of the poem will accumulate meanings beyond the author’s intention.
Finally, as Ankersmit argues in *Sublime Historical Experience*, the goal of nearly all art is not simply to transmit propositions about the world but to give us an experience of making and receiving these meanings: art is a drama of consciousness, not a drama of empirical knowledge. Accute attention to aesthetic experience is still a prerequisite, to me, for any critic.
"The question of whether you wish to interpret something does not depend on whether you see intention behind it; it depends on whether you find it interesting.”
I agree, but in saying “The moment you decide that nature caused the effect, you will have lost all interest in interpreting the formation, because you no longer believe that it has been produced intentionally, and therefore you no longer believe that it’s a word, a bearer of meaning,” isn’t Fish acknowleding this implicitly? You no longer have “interest"--it’s no longer interesting--because you can’t see any useful purpose in fooling with it. It has no obvious consequences. (It may have less obvious consequences if you choose to regard interpretation of such a thing as a game, but this doesn’t obviate the fact that in the ordinary circumstances Fish is dicussing “interpreting the formation” doesn’t get you anywhere.)
Let me bring out some of the implications of the existence of “Cent Mille Milliards de poemes” for this discussion.
The book consists of ten pages. Each page has been cut horizontally into 14 strips. On each strip is printed a line from a sonnet. The lines are so constructed that each of the first lines may be combined with any of the second lines and any of the third lines and so on, to create a sonnet. The rhyme scheme is preserved. Each of the ten first(second, third, . . . fourteenth, respectively) lines follows the same grammatical structure. The lines are written in a style which is sufficiently abstract (lofty?, vague? portentious?) so that whichever of them is chosen, the semantics will not clash.
The result is there are 10^14 possible sonnets, a hundred thousand American billion. Each of which is formally correct, syntactically correct and semantically consistent. By choosing to fold back some number (between 0 and 9) of strips at each line position, the reader chooses a sonnet to read. The likelihood is that this is the first time that sonnet has been chosen to be read (the book was published in 1961 in a small edition; it hasn’t to my knowledge been reprinted; Amazon France shows four copies, _neuf_ or _comme neuf_ available for upwards of $40 apiece. Few people have chosen to read any of the sonnets and there are many possible), not excluding by Queneau himself. That sonnet has a meaning. The words and sentences in that sonnet mean something. Queneau did not intend that meaning. Queneau had a general intention that all of the possible sonnets would have some meaning, and worked very hard on the style of the lines to achieve this, but he had no intention as to the meaning of any specific sonnet. Indeed, he would probably have been surprised by the meaning of any individual sonnet, were he to have read it.
Peter Sattler above: “But the fact remains that when we try interpret what utterance X means, we are always reduced to making educated guess about what it’s author meant by that utterance.”
But this example shows that isn’t true. It may be a good heuristic for many utterances. But it breaks down here. Queneau doesn’t mean by any specific poem. You can’t make an educated guess about what Queneau meant by it.
We have to be very careful about the use of the word “intention.” “Cent Mille Milliard de poemes” has been produced intentionally. Queneau had intentions about the book as a whole. But the individual sonnets that one might read . . .: No. There is a difference between a text and the work that contains it. We can ascribe intention to the work without ascribing intention to any particular text within the work.
I’ll just make a quick response to Peter Sattler, above, even though I haven’t really digested his full comment or gotten on to the others yet. He misunderstands one thing that’s quite important. “You say that you add in just a pinch of “as if” authorship. (What might an author have meant had he or she uttered such-and-such a poem?) But, in my eyes, that pinch is enough.”
Response: No. I don’t admit to helping myself to any ‘as if’ authorship. I admitted that one COULD take a pinch. Thinking of ways to unify the poem by postulating an author is one way you could go. I thought I needed to grant this, lest people accuse me of not seeing it. But the crucial point is that this element is optional. (And now I’ll read Peter and the rest of you more carefully before saying more.)
But “Cent Mille Milliards de poemes” is a toy, like the Mad Lib. Such things are going to become more and more common as the technology for randomly producing grammatical text gets better. I think that this unforgiving technological improvement is going to at some point drive literary studies to have to admit, once again, that some texts really are better than others, and that authorial intention really does have practical importance. Borges went through all of this in his story about the library containing a book for each possible combination of letters. It is possible for an automated process to produce artistic text, but it is very unlikely.
Imagine that the waves on the beach didn’t form letters as a sort of rare phenomenon—imagine that every wave you ever saw randomly formed letters. Or, contrariwise, imagine that all sculpture was made out of fluffy cotton, so that it looked like clouds, and so that all fluffy clouds looked like examples of random sculpture. People would start to get very annoyed with those who kept pointing to waves or clouds and saying “Look at that! Isn’t that just as good as anything an artist could have done?”
In fact, what’s going to happen is that people are just going to start ignoring, for purposes of artistic evaluation, any text without a certain degree of narrativity or coherence. It won’t really matter whether that particular text was actually produced by a program, or just so poorly written that it might as well have been written by a program; people will not see it as being worth their time in either case. And if the programs ever get good enough to produce a work that really can’t be distinguished from a human’s, people will either evaluate the intention of the programmer, or the programs will have passed the Turing test, which is as good a definition of AI as any.
John and Belle’s Wordsworth collaboration reminded me of Lewis Carroll. Games and riddles....
“Cent Mille Milliards de poemes” may be describable as a toy - like Julio Cortazar’s “Hopscotch” - but it is also modernist artefact, isn’t it. The issue of the artist’s intention - his sincerity - is exactly what’s at stake in works that swing on an element of chance and hazard, same as in works like Carl Andre’s Equivalent VIII; the occasion, I should think, of many, many deeply annoying conversations in brickyards, on building sites etc. I like Rich’s point that the status and weight of intentionality in a given work is more or less dependent on historical (institutional?) factors.
Peter Sattler: are you left-handed perchance? I have that left-right word confusion too.
Rereading my piece, I think the tone didn’t come out right. In pouring 10,000 words over an argument which, I say, has been refuted by Wilson and Searle a decade ago, I end up sounding a bit ostentatiously disdainful, when I hope I was going more for a ‘yes, this has been said before, but we’re getting somewhere new’ kind of a vibe: the Knapp and Michaels’ argument is clear and the conclusion provocatively simple - but wrong (I say); so not a half bad starting place for sorting out conceptual issues and such. (Clear and simple and wrong is not a bad score in philosophy.) Now, a couple substantive points.
First, I think I failed to come to clear grips with Knapp and Michaels’ odd usage of ‘theory’. The problem is that they define it functionally: to ‘do theory’ would be to have significant effects of a certain sort. I think that’s too confusing, because you can’t very well let go the old sense of ‘theory’: a general, abstract and/or explanatory account. (I should have stated this more briefly, rather than getting distracted by the prospect of kicking Culler in the shins.)
Second, I make the point that Knapp and Michaels don’t seem to consider that categories like word and sentence are vague, family-resemblance affairs. Messy mix of intention and formalism. Their argument seems to hinge on his not being the case. Do they have a comeback?
Now I’ll respond to some of Peter Sattler’s points:
Dictionaries and resemblances and the predictable patterns of English usage can (often) help us to interpret such utterances. But the fact remains that when we try interpret what utterance X means, we are always reduced to making educated guess about what it’s author meant by that utterance.
But this surely isn’t true of artificial languages like chess notation. I realize this may seem off-point when we are discussing poetry and literature, natural language products; but the argument isn’t about poetry and literature or natural language; it’s about the very nature of language. So any counter-example, in any language, refutes the argument.
I admit, I have all but ignored your focus on “implication” and “significance.” I just don’t see what those aspects of meaning have to do with the question at hand – the meanings of particular utterances.
In a certain scene of a certain play, the fact that Hamlet killed Polonius means he has to hide the body or it will be discovered. You can’t possibly understand the scene unless you understand that. So you can’t possibly interpret it right if you don’t. But this meaning - an implication, generating a significant dramatic intention - is a function of neither sentence meaning nor author intention. It’s a function of the fact that dead bodies are large and people are shocked to find them in the queen’s bedroom. Also, Polonius was an important guy, and killing important guys, even if you are an important guy ... has implications. (There are intentions of human being’s getting in here, but not Shakespeare’s, so far as I can see.) So the fact that implication and significance may not affect the meaning of sentences does not mean they are dispensable for interpretive purposes.
You say that you add in just a pinch of “as if” authorship.
Responded to this already. I said I could, not that I have to.
But do we still want to say that the Chinese utterances made by the man in the box have the meanings that the people outside the box are ascribing to them? The boxed man writes squiggle-squiggle-squoggle –- marks that usually mean, “I understand Chinese” or “Help, I’m trapped in a box” or “I want a pizza.” But do we want to say that these particular utterances are meaningful in this “standard” (or even non-standard) ways?
After all, taking the first example, the boxed man does NOT understand Chinese. Neither did he intend to lie about that fact. In fact, he didn’t intend to communicate any meaning at all when he made those squiggly marks. If the people outside the box interpret those marks to mean that the man inside the box understands Chinese (see utterance one, above), then they are mistaken. Understandably mistaken—but mistaken, nonetheless.
You are mixing up senses of meaning again, and mixing meaning and truth. They may take the fact that the words mean things to mean (in the sense of imply) that there is someone in the box who understands. But then they are simply wrong. The slip says ‘there’s a man in this box who understand Chinese’. They understand what that means (sentence meaning). What they are mistaken about is its truth. What the sentence says isn’t true.
More importantly, have any of you ever hit a homerun in a regulation baseball game?
Does Little League count?
Yes, yes—little league, of course. Okay, point taken.
I appreciate your thoughtful reply. I understand the nature of our disagreement much better now. To close my end of things, please allow me to shore up my ideas on one topic: the Chinese Room and the limits of sentence meaning. (I do so because I think I was sloppy in my previous post.)
Personally, I have no beef with the existence and utility of “sentence meaning.” I just don’t think such meanings matter in any final or even definitive way. Sentence meanings, as I see it, are simply what “most people” mean when they make certain types of marks or noises in certain languages. They are an expression of common communicative practices.
It is certainly possible, therefore, that I could make doodles that look a lot like Chinese -– so much so that they could be read by a person literate in that language. And sure, those marks might have a “sentence meaning” in the basic sense that when people make such marks, they are usually intending to mean such-and-such. The same goes for the man in Searle’s Chinese Room.
But I still can’t go much beyond this, and I’m not certain why we need to.
First of all, when the readers of my marks are interpreting them, they are not just trying to figure out what people USUALLY mean when they say such-and-such; they want to know what *I* mean here and now. And when they find out that I don’t know a word of Chinese, they would probably not have much more to say than, “Weird, because these marks look just like Chinese words, and they usually mean such-and-such.”
To continue in this vein, let’s draw out another simple example, inspired by the AnalPhilosopher post. How would the reader of my Chinese marks –- based just on “sentence meaning” –- know whether my particular utterance was intended ironically, deceptively, figuratively, etc? How would they know without trying to grasp to my particular intentions? (Stay with me a bit.)
Yes, I doodled out the Chinese sentence, “I’m really an expert in the Chinese language.” Most of the time, when people write such a sentence, they mean that they are experts in the Chinese language. But sometimes people don’t mean this at all; sometimes they are being facetious or flippant or ironic or punning or figurative. Which one applies to my doodle? Does my utterance possess the standard meaning or the non-standard meaning, the serious meaning or the “playful” meaning, the literal meaning or the figurative meaning?
Was I being ironic when I scribbled it out? Am I being ironic now? Could I produce an ironic utterance without intending to?
I suppose one could say that, without intention, all utterances revert to the standard (non-ironic) sentence meaning. But that just goes back to averages –- to what “most people” mean most of the time.
Further, if you need to grasp my intentions to know if a sentence has an ironic or figurative meaning, wouldn’t you also need to make a claim about my intentions to say that the utterance is non-ironic or literal? To say that standard “sentence meanings” take the day unless extra ironic (speaker) intentions are added seems to beg the question in the extreme.
Thanks for bearing with me and for a great discussion, John. (He said, without a hint of irony, sticking purely to the sentence meaning.)
Peter says, “Sentence meanings, as I see it, are simply what ‘most people’ mean when they make certain types of marks or noises in certain languages.”
This is certainly a natural and intuitive idea. Unfortunately, I don’t think it works. George Wilson argues for this point in his paper. Consider the sentence ‘Ari’s sister Kayla likes his running shoes more than she likes elephants.’
Suppose that awhile ago some kids were hanging out and decided to do something fun. They decided that whenever one of them uttered that sentence, he would mean by it: some enemies are approaching. Since these kids imagined themselves to have lots of enemies, they uttered this sentence many times, and each time meant by it: some enemies are approaching. So most of the time that sentence has been uttered by someone, he has meant by it: some enemies are approaching. So, Peter, if your view of sentence meaning is correct that sentence means (i.e., its sentence meaning is) that some enemies are approaching.
However, that is clearly wrong (right?). Even if a while ago some kids did what I described, the sentence ‘Ari’s sister Kayla likes his running shoes more than she likes elephants’ still means that Ari’s sister Kayla likes his running shoes more than she likes elephants.
I’m not sure if it matters to any of your overall points that your view of sentence meaning seems wrong, but I thought I’d point it out anyway. Also, if you’re not entirely convinced by this argument (I don’t think I am yet), then perhaps you will take it on authority that your view of sentence meaning (not that it’s some well thought-out considered view of yours) is false: George Wilson says in his paper that it is a commonplace in the linguistics literature that it is false.
Hmm. I guess you’re saying that sentence meaning is just the plain old, what-it-says meaning. Can these meanings change? How about if most speakers adopt the change?
For example, what’s the (current) sentence meaning of “The chickens are coming home to roost” (or “We’ll be back when the cows come home” or “The shit is going to hit the fan” or “Fat chance")?
Taking the first as an example, I bet it used to mean that, well, the chicken were coming home to roost. But if you used that “sentence meaning” as an interpretive guide these days, you would frequently find yourself in error, fruitlessly searching the landscape for returning poultry. Hasn’t the standard meaning –- the sentence meaning—shifted?
And that’s why I defined the phrase “sentence meaning” in such a loose, best-odds, “most people” way. Because I think it only works as a loose, best-odds, most-people concept.
Well, I do think that the meaning of a sentence can change. However, what this seems to show is that the meaning of a sentence *somehow* depends on what human beings do. What this does not show is that the meaning of a sentence depends on what human beings do in the particular way you say it does. That is, it does not show that the meaning of a sentence is a function of what most people mean by it.
John: some thoughts about your very interesting post.
Let’s agree that K&M are/were confused about sentence v. speaker meaning, and that this undermines the letter of “Against Theory”, as written. The question is where this leaves Searle, and where it leaves us.
I continue to think that Searle’s article gives up the game. Yes, sentence meaning is independent of the speaker/writer’s intention. But if sentence meaning is the ONLY kind of meaning for which intentionalism fails, then intentionalism is true of everything literary critics most care about. So K&M are fundamentally right. You reply in two ways.
(1) “The difference between saying something would be odd and saying it would be impossible is rather large.”
Well, sometimes it is, sometimes it isn’t. Our question is whether intentionalism is true of the object(s) of literary-critical interpretation. K&M overshoot in saying that it is true of ALL meaning. Searle points out that sentence meaning is an exception. But if literary-critical interpretation is not primarily concerned with sentence meaning, but with meaning(s) of a different kind, this is a small concession for them to make.
(2) You argue that, once we distinguish speaker meaning and sentence meaning, K&M have no ARGUMENT for intentionalism. All that is left is assertion (the fortune cookie).
Again, I’m not convinced, for two reasons. First, if you believe in Searle’s view, his article actually provides the missing argument.
P1: There are only two kinds of meaning: speaker meaning and sentence meaning.
P2: The object(s) of literary-critical interpretation are meanings, but not primarily sentence meanings.
P3: Speaker meaning is (roughly) identified with speaker/writer’s intention.
Conclusion: intentionalism is true of the primary objects of literary-critical interpretation.
The premises are controversial, but you can see the bones of a valid argument here. It could reasonably be described as an improved version of the argument of “Against Theory”. That’s why I keep insisting that Searle merely qualifies the view of K&M, and does not really reject it. (I don’t mean that I accept the argument above.)
Second, I think there are materials in “Against Theory” for a different, and perhaps better argument for intentionalism about the object(s) of literary-critical interpretation. But that’s a topic for another place.
as I mentioned when you made a similar point in the thread following my post, my reaction was a less articulate version of yours. But, if it’s true that Searle offers but an ultimately insignificant qualification to K and M, that leaves the question of why K and M (along with Fish) and Searle both think otherwise.
It could be that they’re misled by a battle of the egos or a kind of institutional conflict between lit and philosophy, but that seems unlikely--and an ungenerous view. More likely both think they have the best account of language and are eager to see the other concede to it. In his San Diego law review article, though, Fish stresses something that I think is true for K and M as well: that to their understanding their argument isn’t limited to literary-critical interpretation at all, but a general account of linguistic meaning and one with what might be called ethical stakes. I.e., that it’s bad for people and society to pretend that they’re interpreting when they’re indifferent to what a speaker means. That’s really the gist of Fish’s op-ed too.
Arguably, you could say that a concession about “sentence meaning” wouldn’t hurt this view either--that, e.g., interpreting law really has the same features as interpreting literature and likewise requires attention to speaker meaning. But leaving open the possibility of “speaker meaning” would also leave open the possibility that that’s not so--and thus potentially to Scalia’s textualism.
But Sean, what about the point made awhile back in the Fish-post comments to the effect that laws possess nothing we could call “intentions,” that they are hammered out by a series of competing interests in order to find the least offensive position on a certain topic. Is the intention behind a law the sum of the intentions of all the various statespeople who wrote and passed the law?
If interpretation is dedicated to “finding the speaker’s or writer’s intended meanings,” then its mission is, while noble, also impossible. All we’ll ever really have recourse to is the text (and perhaps some minimal contextual information), and all we can ever be sure of is that some authorial position—whether an individual or a group—chose to put those words in that order. An appeal to intention can’t settle a debate, except insofar as we can define the limits of an author’s consciousness with its historical limits—i.e., Michaels’ example that we must believe that Marvell’s poem is written in 17th century English (’tho even there he gets his OED wrong: “vegetable love” in the 17th century could mean both “slowly growing” as well as “of or like a vegetable,” and so this slowly growing “love” could also be, say, cucumber- or carrot-like, if ya know what I mean).
It seems to come down to the meaning of the word “interpretation.” From Michaels’ stance, what historicists generally do cannot be called interpretation. But most lit scholars, for instance, tend to use the word interchangeably with “analysis,” “close reading,” “historicizing,” and so on. Seeing as Michaels himself doesn’t “interpret” literature in this sense in his own work, we can assume that he doesn’t want to argue that the true task of literary criticism is interpretation. If anything, *Our America* displays an almost structuralist grounding mixed with some Foucauldian discourse analysis.
(There’s also the “what’s the right word” problem, as evidenced on last night’s episode of *So You Want to Be a Hilton?*. If meaning is the same as intention, why do we struggle to express our meanings in words and why are most artists—and married people—dissatisfied with the way their words expressed their intentions—when they even write in such a way? Meanings are part of experience, are private, while language is social, conventional, etc. There’s also the denotation/connotation problem: unless we have access to the author’s mind, we can never really know what’s connoted by his/her language. All we can know is what’s denoted, and then we’re back to sentence meaning. Imagine a poem written in English about how beautiful vultures are, but by an author from a culture where vultures aren’t equated with death and rotten meat: the author’s familiarity with English *could* mean s/he is familiar with English cultural associations, but we can never know. Which means the poem could be ironic or could simply be a straight-ahead “vultures are pretty” poem. We can never know. The fact that we fall back on sentence meaning is seen in Michaels’ example from Andrew Marvell, in which he appeals to dictionary meanings and not to, say, biographical information.)
So while I see what’s at stake for intentionalism in legal analysis, I don’t quite see what difference intentionalism makes in literary analysis (besides restricting our use of “interpretation” to “arguing for the author’s intended meaning").
Luther, you continue to make the same mistake over and over again. Other people’s capacity to understand the meaning of my utterance has no bearing on what the meaning of my utterance is. My dissatisfaction with the degree to which others understand my meaning through my utterance does not mean that my utterance did not mean what I meant by it.
Put it another way: Michaels/Knapp are making an argument about what meaning is. You think that the epistemelogical problem of knowing meaning has some bearing on what meaning is. But that’s wrong. You don’t have to read an authors mind to be persuing an author’s intention. You just have to be interested in what the poem means.
That’s also why you are wrong about WBM’s dissatisfaction with historicism. Being interested in meaning (intention) does not automatically dictate a method for finding it (e.g. mindreading). Nor does it rule out the possibility that one way to go about finding intentions is to look at the constraints that historical formations or discourses have on intentions.
Luther, I hate to pile on with the same old objection, but “If interpretation is dedicated to “finding the speaker’s or writer’s intended meanings,” then its mission is, while noble, also impossible. All we’ll ever really have recourse to is the text (and perhaps some minimal contextual information)” is just flatly false as a general proposition. It may be true for laws written centuries ago, and I can’t quite tell if you’re trying to continue the context of your first paragraph. But it just isn’t true in general: living authors exist. Even dead authors who supplied a good deal more than minimal contextual information while they were alive exist.
I don’t think that interpretation should be dedicated to finding a speaker’s or writer’s intended meaning, that sounds boring. But I don’t think that it can ignore it either, that sounds trivial. A good interpretation has to take one’s best guess at intended meaning into account and go beyond it.
I’ll make a flatly humanist (therefore ideological) statement of my position: since people create meaning, you have to allow for a role for every person involved—the author, the reader, the critic—in creating the meaning of the text.
HZ: You argue, “Other people’s capacity to understand the meaning of my utterance has no bearing on what the meaning of my utterance is. My dissatisfaction with the degree to which others understand my meaning through my utterance does not mean that my utterance did not mean what I meant by it.”
That’s patently absurd to me. The intention behind one’s words is a necessary but not a sufficient condition of their meaningfulness. The failure of a speaker to “say what he means” is not a failure of the listener; the utterance doesn’t “mean what he meant” but rather “means what he says.” An example:
Boy: Hey, bitch, give me some roast beef.
Girl: Screw you, sexist boy.
Boy: Oh, no, don’t misunderstand me. I was just ironically imitating gangsta rapper discourse.
Here, the girl isn’t “failing to understand the meaning of the utterance.” She understands it as well as she can. The boy’s remark’s lack of proper irony-signaling devices meant that his intention never *made* it into language. The language got beyond him. His code obscured his message. That’s not a problem with the receiver; it’s a problem with how the sender encoded the message, and so in terms of human agency, it’s a problem with the sender, not the receiver. Meaning is the product of the complex interactions of the entire act of communication. I’ve said something meaningful when someone understands me, and not until then (or, if a book falls in a forest, it doesn’t make a meaning). I can’t know I’ve “made meaning” until someone else has decoded my message. Until then, I could be speaking in incomprehensible grunts for all I know. My intentions might be good, but my message, my code, my contact, all might be screwed up. And that’s my problem, not my receiver’s.
The epistemological argument I raised earlier only serves to ask this: why posit the existence of something—intention—we have no access to? Aren’t we just back in Kantian “noumena” territory? And, as Hegel argued, how can we know something exists without being about to know it? Either we know it or we don’t. If we can’t ever really know it, we should just proceed without bothering about it. Rather than think that the poem’s meaning is this noumenal intentionality, we might as well proceed without it.
Rich objects that we *do* have some access to intention, via author’s statements. But with intention, as with all essences, we can’t have “limited access” but either total or no access, because with intentions, it’s either all or nothing. Either the subject has complete self-presence of consciousness, or the subject interprets his or her own discourse as much as anyone else. If I intend to ask for the salt and instead say, “Can I please have an assault?”, knowledge of my conscious intention will get you nowhere.
Or, to use an example I’ve used before, let’s consider Katherine Mansfield’s “The Garden- Party.” In a letter to William Gerhardi (the “English Chekhov”), Mansfield wrote: “And yes, that is what I tried to portray in ‘The Garden-Party.’ The diversity of life and how we try to fit in everything, Death included . . . [Laura, the young protagonist] feels things ought to happen differently. First one and then another. But life isn’t like that. We haven’t the ordering of it.” All right. It’s a good place to *begin* thinking about the story, until we realize that this statement of intention says little to distinguish the meaning of the story from a gazillion other stories, poems, plays, films, comic books, and so on. And then we also see that the statement begins with an agreement, and we begin to suspect if Mansfield is simply allowing Gerhardi to believe he’s more intimate with her stories than he is. That the story is all about class and guilt, and that Mansfield doesn’t mention class or guilt at all here, suggests that the way intentions play out in signs, images, plots, and so on is so complex as to put us on far safer ground to pay attention to the text and not to this *other* text – Mansfield’s letter – a text the intentions of which we would then have to interpret as well (meaning: we’ve only increased our text, not strengthened our interpretation). Even if Mansfield intended to say nothing about class, the fact that she plays out the drama between life’s chaos and the human desire for order in terms of a poor laborer’s death occurring in the midst of a gorgeous bourgie garden party says as much about the text’s *meaning* as does her statement of intention.
HZ also argues, “You don’t have to read an authors mind to be persuing an author’s intention. You just have to be interested in what the poem means.” If intention is meaning, this is simply a tautology. To be interested in meaning is to be interested in what an author meant, and to be interested in what an author meant is to be interested in a text’s meaning. It doesn’t seem like we’ve said anything about either meaning or intention by equating the two. What we *have* done – and this goes back to my memories of *Against Theory*—is to try to show the aims of theory to be senseless.
Rich, I agree that we should use everything possible to shed light on what a text means. This means exploring our own experiences for connotations as well as exploring the author for hints. But the text still feels like the surest ground we’ve got.
(Finally, I suggest taking a look at Ankersmit’s recent *Sublime Historical Experience* for an excellent take on experience without a subject – a phenomenon that creates problems for Michaels’ other ideas in *The Shape of the Signifier*.)
"The intention behind one’s words is a necessary but not a sufficient condition of their meaningfulness. The failure of a speaker to “say what he means” is not a failure of the listener; the utterance doesn’t “mean what he meant” but rather “means what he says.””
The second sentence contradicts the first. If you claim that my ironic sentence can _fail_ of its meaning simply by being misunderstood, then you are claiming that intention is not even a _necessary_ condition of meaningfulness. Indeed, what you are saying is _you don’t care what the sentence means_.
Your reasons for that indifference oscillate between A) an epistemelogical difficulty (we can’t ever know for sure, so let’s just do something else and call it meaning) and B) An ontological claim about meaning (meaning is a “complex interaction” between a speaker’s intention and an experiencer of speech). Of course, B contradicts A in precisely the same way as those two quoted sentences contradict each other. If we have decided that intentions are “noumenal” (really, Kant is not helping you here, but there you are) and therefore not up for consideration, then intentions are in “complex interaction” with nothing-- they are off the table. And if intentions are in complex interaction with the experience of listeners, producing what you call “meaning” (but which would be better called “communication") then they are not, in principle, inaccessible even if they are, in practice, unknown.
I have no idea what the payoff of the claim that meaning is intention does not “say anything about either” is meant to be. It says that to seek one is to seek the other. It is true that it doesn’t tell you how. The Mansfield example is fine-- the claim that meaning is intention does not limit interpretation to the tracking of _statements_ of intention. I might, for example believe that Mansfield had _unconscious_ intentions to write about class and guilt. Or that historical forces were at work in such a way that her overt intentions masked a set of more pernicious ideology. In each case, I would marshall quite different kinds of evidence for my interpretation. But what my interpretation would be an interpretation of is what Mansfield meant by The Garden-Party.”
I don’t see the contradiction between: (a) seeing intention as a necessary condition for meaning; and (b)seeing the failure to “say what you mean” as resulting in a misunderstanding on the receiver’s part (instead of a mis-communication on the sender’s part).
Statement “a” speaks to the Wimsatt and Beardsley idea that poems come out of heads, not out of hats (’tho John Holbo has problematized even that). I agree with Michaels insofar as he simply means that statements, to have meaning, must have been intended. But to my understanding, intention at best means “intention to mean” or “intention to communicate”—intention doesn’t define the total act of communication. If I intend to build a chair but instead make something that looks exactly like a paper hat, it’s not your fault if you think it’s a paper hat. My intention in doing anything can’t completely define an action. Not meaning to offend someone with my words doesn’t mean that my audience is “misunderstanding” me when they take offense. My point is simply that, in the example above (i.e., “Hey bitch give me some roast beef"), unless the communication act involved clear social conventions to signal irony, the woman hasn’t *misunderstood* the boy; the boy has miscommunicated with the girl. The girl would have had to read the boy’s mind in order to know that he was being ironic.
Secondly, I didn’t want to argue that intention *is* Kant’s noumena, but rather that, just as Hegel avoided the problem of the thing-in-itself by asking how can we know there’s something we don’t know without knowing it. But you’re right, it just complicates things. My bad, dog.
I also disagree that the next two claims are contradictory: (a) we don’t have access to the full intentions behind any act of communication, and so it’s easier just not to worry about intention; and (b) meaning is the product of a complex interaction between sender, receiver, message, code, contact, and context. I’m not saying that meaning is an effect without a cause in intention; I’m only saying that unlike the other aspects of the communication model, we only know the intention by its effects (even statements of intention are effects, as an author tries to express her intentions in another way than the poem or novel or whatever). Further communication can clarify the intentions behind a previous utterance, but they don’t really change the meaning of it in the moment because the context will have changed, and so we’ll be dealing with a separate act of communication.
Finally, HZ writes, “Or that historical forces were at work in such a way that her overt intentions masked a set of more pernicious ideology. In each case, I would marshall quite different kinds of evidence for my interpretation. But what my interpretation would be an interpretation of is what Mansfield meant by ‘The Garden-Party.’”
Well, now we’ve got a really weird idea of both “Mansfield” and “meant” in that final sentence. Mansfield can’t mean something that historical forces or ideological pressures are doing to her or her composition. It would seem that intentions, to be intentions of the subject, need to be conscious. What’s doing the meaning here is at stake. Is it the it or the I? If, in the old (post)structuralist chestnut, language (or ideology or the unconscious-as-language or whatever) speaks the subject, I don’t quite know what “intention” means anymore. But perhaps I’m misunderstanding something here. It’s late.
1) You are treating intention to mean and intention to communicate as synonyms, or at least close enough to fall on either side of an indifferent or. That is precisely wrong. You can mean something by your marks without having any intention to communicate.
Note that your example in paragraph 2 are not examples of signification: the act of chairmaking is satisfied by meeting the criteria for making a chair, and it is not at all clear that one of those criteria is that someone else _think_ that I have made a chair. It may not be your _fault_ that you think you have seen a hat rather than a chair [what does “fault” even mean here?] But this is precisely parallel to distinction between intention to mean and intention to communicate. I intend to make a chair. No one has yet suggested what the formal features of a chair are, and I am so far indifferent to the question. But can you not see that if I intend for _you_ to acknowledge my chair as such, that is an additional intention?])
Thus it is correct to say that intention does not define the act of communication. Intention defines the act of _meaning._
2) There is nothing in the claim that meaning=intention that required intention to be conscious.
1) “You can mean something by your marks without having any intention to communicate.” Not quite. Unless you yourself act as the receiver of your own marks (and let’s remember that Michaels talks always about signs, for marks cannot mean). Until one communicates, one can’t ever know if one has produced meaning. One might be speaking in grunts or incomprehensible scat sounds. My intention might be to say “All cats are grey,” but what I might say is “Old cats are grey.” This utterance doesn’t “mean” the former, even if that’s what I meant to say. Or else we’re in the land of mind-reading.
2) You’re right about the difference between making a chair and making a statement, except that there are *social* conventions for making both chairs and meanings. Meaning cannot be reduced to what goes on in a brain, because what goes on in a brain is: (a) infinitely complex; (b) private; (c) unique; (d) unrepeatable. Whereas what goes on in an utterance or any use of signs is: (a) limited by convention; (b) social; (c) not unique; (d) iterable to the point of absurdity.
(3) Finally, to be able to say something like “Katherine Mansfield intended meaning x,” we’re either talking about conscious intention or we’re not talking about Katherine Mansfield. This is why Freud distinguished between an I and an it. I’m being a bit ridiculous here, but it seems that for *me* to intend something, I have to be in control of the act of intending. Otherwise, something else is intending—the unconscious, history, or whatever. The unconscious doesn’t intend—it doesn’t express, it indicates, it produces symptoms. My Freudian slip doesn’t “express the intention” of the unconscious; it is as much a symptom of the unconscious as my cough is a symptom of my bronchial infection. Same with the effects of history on a text: history doesn’t “intend,” anymore than the waves that erode a rock formation into the shape of “Help!” intend.
So either history and the unconscious can’t effect meaning as meaning (because what I intended exists as a moment in brain, even if nowhere else?), or meaning can’t be reduced to intention. To say that an author “intended” what seems to be the result of historical or unconscious forces would be to argue that the author recognized these aspects of her text, made them conscious, and accepted them as part of her intention (as she would an editor’s revisions).
(4) In any case, this argument is going nowhere. Neither of us is going to change our minds.
Luther B said: “My intention might be to say “All cats are grey,” but what I might say is “Old cats are grey.” This utterance doesn’t “mean” the former, even if that’s what I meant to say. Or else we’re in the land of mind-reading.”
I promised myself I had left this debate, but just when I thought I was out . . . they pull me back in. I’ll leave it to HZ to continue the discussion of epistemology and various kinds/sources of intention. Suffice to say, I think he and I are in agreement.
And I think we will agree in saying that, perhaps counterintuitively, your story both makes the case for intentional meaning and shows the limits of so-called sentence meaning.
Let’s go to the cat’s tale. Okay, perhaps I have said something that sounds like “Old cats are grey” –- because of a slip of the lip, or a heavy Southern drawl, or a piss-poor English vocabulary. But if I intended for that utterance to mean that “all cats are grey,” then that’s what that particular utterance meant.
Are we in the land of mind-reading? No. But the fact that I communicated in an unclear, or abnormal, or substandard fashion does not change the (speaker) meaning of my utterance. Yes, I used a string of words that is usually uttered to make claims about “old” cats. But this simply acknowledges a statistical fact about English –- namely, that when English speakers say those words, they usually intend to mean such-and-such. And listeners use these so-called sentence meanings as a guide for interpreting and understanding the meanings of new utterances. And, yes, those guides usually work, passably. They make language function as a communicative tool.
But sometimes –- and Davidson would say a lot of the time -– people use language in strange, abnormal, non-standard ways. And at those times, sentence meanings only help me partly and non-definitively to interpret an utterance. In fact, sometimes they don’t help much at all.
A case in point, which I have mentioned already: irony. I can use the sentence, “Paris Hilton sure is a great role model of young girls,” to mean that she is a great role model (if I’m speaking literally) or that she is clearly not a great role model (if I’m speaking ironically). The meaning of my particular utterance is determined by my intention.
The fact that the “sentence meaning” seems to guide you in one direction doesn’t affect the meaning of my utterance. (I am intending to violate the standard sentence meaning.) And the fact that you could misunderstand my statement –- assuming that I’m being straight, when I’m being ironic (or vice versa) –- doesn’t affect its meaning either.
So, yes, you can say “old” to mean “all”; you can say “is great” to mean “is not great”; you can say “black” to mean “white,” “up” to mean “down,” and “left” to mean “right.” And your utterances will mean what you intend.
We do these things all the time, especially in literature. It’s not always a great tactic if you want to make yourself understood. But that’s language too.
Now I’m really out of here!
A Freudian slip is when you say one thing and mean your mother.
To put it another way: why suppose that intention is univocal?
Why, also, suppose that intention strictly precedes utterance, and doesn’t to some degree coincide with it ("how can I know what I think until I see what I say?") or even arrive after the fact ("what I *think* I was trying to say was...").
Intention can also get lost along the way - I’m sorry, I have no idea what I’m trying to say here.
If I say something very insulting about a person’s mother, and word comes to them of the outrage and they ask me what I meant by it, the chances are that they don’t want me to talk about whether I meant the word “prostitute” literally or metaphorically. If I start pontificating airily about how we are all, in some sense, whores then my antagonist will most likely suppose that I’ve taken note of his or her superior size and musculature and decided to back down from the aggressive posture established by my initial speech act.
As I think Derrida was the first person to notice, interlocutors sometimes die. Their utterances, especially written ones, often hang around after they’re gone, in books and graffiti on toilet doors and suchlike. Sometimes they get quoted by other people, in some cases people who have no idea who said them first.
Wasn’t it Shaw who said that the iterability of the graphic mark was both the condition of possibility of its being re-marked in an interpretative context that might lead back to an original intention, and the condition of impossibility of finally confining that mark to that context (since it can always crop up elsewhere, beyond the horizon of any finite contextual delimitation)? He probably put it better; or it may have been Wilde.
Does it make any difference if the language being inscribed by the homunculous in Searle’s box is Klingon? Or COBOL? Is the meaning of computer source code identical with the author’s intention?
I meant, I think, homunculus.
More seriously (in tone at least):
I’m not able to perceive an intrinsic difference between a graphic mark that resembles a word and a graphic mark that is one, even though I might make that distinction in some practical circumstances.
To me, graphic marks that are words just are graphic marks that resemble words, the resemblance being a) in the eye of the beholder, and b) partly conditioned by the beholder’s expectations concerning words and the usual mechanisms by which they come to be inscribed. Graphic marks that look a lot like words but in some other respect disappoint those expectations (for instance if it isn’t plausible that any human agent actually inscribed them) won’t meet the full criteria for recognition of this kind.
When I suppose a sequence of graphic marks to be a sentence, I will also tend to suppose that someone - possibly but not necessarily unknown to me - may have meant something by it.
But those are supposes; insofar as they have a structure, it’s the structure of an interpretative gesture. That gesture may posit, amongst other things, an author and an intention belonging to the author, and will misfire if either is in reality missing, e.g. if the word-formation really is random, or pseudo-random, in spite of my supposing it to be otherwise.
In fact there are several ways in which one could be mistaken about the degree and kind of intentionality involved in an utterance - one might for instance be mistaken about the degree of singular conscious engagement involved in producing a text that turned out to be a mad-lib produced by several parties, none of whom was even particularly paying attention to what they were doing.
If one can be mistaken in several ways, and to several degrees, about intentionality then that implies that one can also be more or less right about it. One question is whether it is the task or duty of (especially literary) interpretation to try to be as right about it as possible. Are there any circumstances in which it might just not matter? Does it matter to varying degrees in varying circumstances, or is it a fixed norm - a line in the sand - for all interpretation everywhere?
My own sensations and recollections connected to writing and thinking about writing suggest to me that having an intention, and inscribing graphic marks that have something to do with that intention, are subjectively quite unlike what I as a reader suppose to have been going on in another person’s head when I try to apprehend the meaning of a bit of ordinary language uttered by them in an ordinary pragmatic setting.
That is, the first-approximation supposes we make as readers about meaning and intention are heuristic, initially reductive and simplifying. They might give interpretation its initial bearings, but the reconstruction of meaning they perform is a simulation (for instrumental purposes, moreover) rather than a direct mirroring of the other person’s psychological reality. Even if we had direct access to that reality, we wouldn’t ordinarily have time to process all the information it contains.
Dennett’s “intentional stance” is partly about making do with limited information, and partly about *deliberately* limiting information to the small range we are usually concerned with. When playing against a chess-playing computer, for example, ,it’s more convenient for me to try to imagine the strategies it’s trying to pursue (even though it hasn’t anything remotely resembling a “psychological reality” that could encompass contents of that sort) than to try to reproduce its computational procedures in my own mind.
Should literary interpretation limit its own range of “supposes” to those which are convenient in such everyday cases? I would answer no, because literature itself does not do so - it can require, and reward, very different levels of attention and ranges of response, and not all of the interpretative strategies it solicits are centred on the divination of authorial intention. One would have to choose not to know that about literature in order to think that, when it came to interpretation, there was necessarily only one game in town.
I believe there are two concepts of “meaning” detached from, but usually communserate with each other. The first is authorial meaning, the second is “third person meaning” i.e third person meaning, the meaning we would assign if we knew nothing of the author. I think arguging over which is right and which is wrong is a bit like arguging over whether up or down is more necessary. I think much the same is true about arguging whether liteary criticism should examine the first or the second “meaning”, I think it should do both, and also add in some intresting but false interpretations for good measure.
By the way, the weird post before me, has that actually got anything to do with our disscusion, or is it spam?
By the way, the weird post before me, has that actually got anything to do with our disscusion, or is it spam?
Obviously spam. In order to get it past the Bayesian spam filters policing this site, the spambot that posted it randomly assembled sections of prose from a variety of sources, thereby giving an impression of intentional utterance sufficiently plausible to convince a Bayesian spam filter, if not an actual human being.
Intention in regards to ordinary discourse or non-literary communication is quite a different matter than the intentions of a novelist or poet, as one might (or might not) infer it from a piece of lit., is it not. Searle, at least from the excerpts I have read, seems to routinely posit mental states--consciousness, more or less--and this includes, or entails intention, and while that may be ultimately necessary (at least until some “field theory” of cognitivism is produced), obviously inferring other people’s intentions and thought processes from their speech is not always an easy task.
Simple commands do work: a parent tells his child “put on your coat because it’s cold.” There is not really much room for doubt, regardless of what postmods might say. The intentions are clear (tho’ perhaps the mental-biological processes leading up to the speech act are not). So the sort of bare bones “external realism” that Searle discusses is not really so much an issue at least at the level of most conversation. Parents in various communities might have different ways of using words to give commands to their children and so forth and that might be sort of Wittgensteinian issue, but there can be little doubt that some sort of intention (following a perception--coldness and the risk to the child) occurs, tho’ that might have been determined in some sense (a need to protect/nurture perhaps?) as well. Searle seems to want to eliminate determinism as a rule, and again, that may not be ultimately correct, once genetic/cognitive pathways are more clearly demarcated. That is to say, the scientific status of intention cannot be inferred ala Descartes or Searle; Searle’s writings are a bit beyond the Cogito but there is still a tendency towards this view of the subjectivity of consciousness which seems provisional and a bit anti-rational. But Searle does claim that our sense of Mind is dependent (completely?) on the brain’s biochemistry and neurology, which is a bit more accurate and worthwhile a step than most lit. critics are willing to commit to.
The celebrated Chinese Room scenario is not necessarily irrefutable: that computers do not currently process syntax in the way conscious minds do doesn’t mean that at no point will a program/CPU/construct be able to (I do think Searle grants that).
Light, in the absence of eyes, illuminates nothing.
I can put together a random word generator, claim it’s generating poems, and a billion Literary Appreciation majors will assign meaning to them. They’ve created something from nothing.
And the Chinese room scenario is just sort of silly; Kurzweil among others has utterly destroyed it. Clearly the man’s “magic answer book” does understand syntax - the scenario merely transfers the agent of understanding from the man to the book.