Fair is foul, and foul is fair

This will be another wonky analytic philosophy post about imaginative resistance. (Here is the last such, with links to others yet further back.) Today I cranked out the first half of an essay - too hastily, although I've been taking notes for the longest time, and the first bit is recycled. It is a bit precious and scholastic, as befits its dignity as an academic paper. It lacks an ending, so don't expect one. But if there's anything fundamentally defective about these feet and torso, it would be helpful if I were informed before screwing on the head. Brian Weatherson, in particular, is instructed to leave critical comments. (And I am sorry that, in the course of the post, I have been compelled to note before god and google that the fiction he has penned of late is hardly of the first water, aesthetically. Brian is a fine fellow. But there you are. Plato dear, truth dearer, yadda, yadda, yadda.)

And I've realized I don't really understand Gendler's paper. That is, it seems to me she should conclude her paper by saying, sensibly: 'so there's really no such thing after all. Just a big misunderstanding.' But she doesn't. I really should email her and request clarification, but haven't yet.

Proceed at own risk. Geek-talk between us academic wonks.

§1

“Fair is foul, and foul is fair”
- Shakespeare, Macbeth (I.i)

In “The Puzzle of Imaginative Resistance” [link to PDF draft], Tamar Gendler sets out to investigate, “the puzzle of explaining our comparative difficulty in imagining fictional worlds that we take to be morally deviant.” Not worlds in which people behave badly (that’s easy); nor ones in which people have deviant beliefs about morality by our lights (again, easy); nor ones in which peculiar conditions make peculiar moral demands of characters; rather, worlds in which, counterfactually, moral truths are different than (we take them to be) in the actual world.

The fact that Gendler’s statement of her puzzle is readily misunderstandable is a prima facie argument for its genuineness. If our imagination has a ‘blind-spot’, one would expect it to be hard to make people aware of that fact. So, by way of making sure we apprehend the contours of the alleged puzzle, consider this sentence Gendler borrows from Kendall Walton:

1) In killing her baby Giselda did the right thing; after all, it was a girl.

Let 1) be the first sentence of a story. Set it alongside these first sentences from Stephen Baxter’s sci-fi novel Flux :

2) Dura woke with a start. There was something wrong. The photons smelled funny.

There is something aesthetically artificial about stopping at the start and asking, ‘now what do you think?’ Standard practice dictates not a pause to take stock but a rush ahead. Nevertheless, let us pause and take stock.

Regarding 2), the first thing one thinks is that in this fictional world there is someone named Dura – who may not be human, but is sufficiently human-like to sleep and be startled – capable of the olfactory detection of photons.

It might, of course, turn out Dura is crazy or hallucinating; or perhaps ‘smelled funny’ is slang for ‘tripped an alarm on the detection equipment’. But the reader probably defaults to the ‘straight’ reading, even though it is objectively the most fantastic. Such, such are the wonders of fantastic fiction.

Turning to 1), we are likely to give about equal weight to three candidate readings. First, the narrator is morally unreliable, i.e. the author has decided to speak in the voice of someone bad, or mad. Second, the narrator is ironic, i.e. half stepping into Giselda’s character by way of informing us she is bad, or mad. Third, the situation is not as it seems, e.g. we will presently learn that Giselda knows any baby girl of hers will instantly be traded for an evil changeling. So, despite appearances, she is doing the right thing even by our lights. There are other possibilities. It might turn out that ‘baby’ means project. (Giselda works in a company in which it is an in-house joke that ‘it’s a girl!’ means ‘it didn’t work out.’ It all goes back to the sad case of Smith, Giselda’s supervisor, who wanted a boy and whose wife had seven girls.)

Not to harp on the fact that first sentences of stories are open-ended affairs, the point is this: in one case – the light-sniffing case - we easily settle into a ‘straight’ reading on which something arguably impossible in the real world is simply and flatly true in the fictional world. In the other case – the baby-killing case - the strictly analogous ‘straight’ reading is so far down the imaginative list that we probably never so much as consider it. It does not readily even occur to us that the author might be envisioning a world unlike our own in this respect: here it is good to kill baby girls.

The obvious response runs: light-sniffing is less strongly and self-evidently impossible than is the rightness of baby girl killing. The former may be a paradigm of something we cannot really imagine (in some sense) but which we can readily conceive (shades of Nagel and ‘what is it like to be a bat?’) The rightness of baby girl killing is just inconceivable. This is why we are more imaginatively resistant to morally deviant worlds. (Please note, for future reference: we are already stepping on our own feet concerning what the ‘imaginatively’ in imaginatively resistant is supposed to mean.)

But the counter-response is almost as obvious. We up the ante from light-sniffing to pure conceptual impossibility – of a non-moral sort – at which point we will allegedly discover that we are more resistant to moral than non-moral conceptual impossibilities. To show this, Gendler authors a charmingly impossible (but morally anodyne) fable, “The Tower of Goldbach”. Here are its bare, narrative bones:

'Mathematicians prove Goldbach’s conjecture: every even the sum of two primes. God is angry at this impertinent lifting of the skirts of bashful Creation to peer at her too-private mystery parts. God falsifies Goldbach’s conjecture by making 12 no longer the sum of two primes, 7 + 5. Humans are upset. If they find 12 righteous men, will God relent? Yes. Seven righteous men, and five, are found, but are unable to sum to 12 righteous men (see above.) More pleading. Solomon intervenes: it would be grandest for 12 to be, and not, the sum of 7 + 5. God agrees. A chorus of 12 and -12 voices hymn God’s immortal glory. The end.'

In the somewhat less abrupt original, the literary atmosphere is (I should say) a whimsical mix of Lewis Carroll and mock-Tolstoyan pietism. As such, it is quite a successful little literary offering. And this is Gendler’s point. There is no imaginative resistance. Quite the contrary, there is imaginative attraction, despite the fact that we are utterly incapable of conceiving how the events in this story could possibly be true. Gendler suggests such a trick could not be pulled off regarding morally deviant impossibilities – or not as easily, or not as well. Hence the puzzle of imaginative resistance. Again: “the puzzle of explaining our comparative difficulty in imagining fictional worlds that we take to be morally deviant.”

Gendler’s paper is a response to an earlier one by Kendall Walton, “Morals in Fiction and Fictional Morality”. He is worth quoting for he throws down the gauntlet with greater force (while admitting that probably this will lead to trouble):

“Why shouldn’t storytellers be allowed to experiment explicitly with worlds of morally different kinds, including ones even they regard as morally obnoxious? There is science fiction; why not morality fiction?

I am sceptical – sceptical about whether fictional worlds can ever differ morally from the real world . . . I have learned never to say never about such things. Writers of fiction are a clever and cantankerous lot who usually manage to do whatever anyone suggests can’t be done, and philosophers are quick with counterexamples. But in this instance counterexamples are surprisingly difficult to come by.”

Thus we may distinguish strong and weak versions of the thesis that there is imaginative resistance. Gendler claims that morally deviant fiction is relatively hard to produce. Walton goes further, speculating that it might be impossible to produce.

In the second half of the paper, starting in §5, I will answer Walton’s challenge directly by adducing six distinct classes of counter-examples. Furthermore, these will not be out-of-the-way cases. We need not colonize barren and forbidding regions of far-flung logical or paralogical space, for there is a wealth of actual, quite ordinary examples. It is not the case that morally deviant fictional worlds are the least bit hard to produce. This suffices to refute Walton. He is just plain wrong all over the place.

Gendler’s case is trickier. It seems to me, given the wealth of counter-examples, the problem of imaginative resistance is basically a pseudo-problem. It seems to me as well that Gendler’s discussion provides her with the basic means to draw this correct conclusion. At the end of her paper she is in a position to say that appearances that there is such a thing as imaginative resistance are misleading. It is all due to factors it would be misleading to label ‘imaginative resistance’. But Gendler does not quite say this, which leaves me wondering whether she neglects to take the last step – or whether I am misunderstanding her. So I will not target Gendler in what follows, apart from some incidental observations along the way. I will answer Walton’s challenge directly. It is quite easy to construct morally deviant fictional worlds. Except that (this being philosophy) it all turns out to be a bit more complicated.

§2

Faith, here's an equivocator, that could swear in both the scales against either scale. - Shakespeare, Macbeth, (III.ii)

Before I lay out my counter-examples, in §5, let me say why a refutation of Walton’s claim may be of some general interest; and let me offer some preliminary reflections that point the way, and smooth the way, for what will follow.

The problem of imaginative resistance – if it were a problem, which I will basically deny – would seem to touch, potentially, on at least six philosophical topics or general areas of interest.

1) Epistemological/metaphysical issues about conceivability and possibility, e.g. does the former ever entail the latter?

2) Semantic puzzles, e.g. how to understand (and model) relations of truth in fiction.

3) The 'principle of poetic license', i.e. for any proposition, there are techniques by which an author can make that proposition true in her fiction. (True or false?)

4) Ethical questions, e.g. does fiction (should it) ethically instruct?

5) Phenomenological questions, i.e. what is it like to be a fiction-reader?

6) Literary critical questions, e.g. how are fictions put together, how do they function, what makes them successful or unsuccessful?

In what follows I have nothing to say about 1 (so far as I can tell), and little to say about 2; mostly I will swing back and forth on the hinge of 3. This will constitute an instructive tour through 4, 5 and 6 – mostly 6. I think the interest of the pseudo-problem of imaginative resistance is that it provokes us to explore a rather intriguing question, which we might not otherwise have asked: how are all the morally deviant fictional worlds we read about in fiction put together? What are the techniques of construction? What are the artistic advantages and disadvantages of working in the medium of deviant morality? (There is a tendency to gesture, rather flabbily, in the direction of infinite imaginative free-play and the divine prerogatives of authors to do as they will. Not that this is wrong, but it is all a bit more involved than the rhetoric suggests. The god must emerge from the machine; instructive, then, to give the machine a tap or two - kick the tires, peek under the hood.)

In the next section let me display the hinge on which we will swing back and forth. The principle of poetic license is equivocal. In a sense it is obviously true, in another sense obviously false. (And sometimes it is hard to to tell which scale – or both - we are swearing into.) And eventually this will lead us to appreciate the converse truth that, in one sense, any morally deviant fictional world you like can be constructed. In another sense, not; in another sense, indeterminate.

§3

“… If the assassination Could trammel up the consequence, and catch With his surcease success; that but this blow Might be the be-all and the end-all here.” - Shakespeare, Macbeth (I.vii)

How is the principle of poetic license ambiguous between a true and false reading? A comparison with painting should help. Consider a candidate principle of painterly license: when it comes to two-dimensional arrays of color, painters can do anything they want. This seems true because, for any area of canvas, any color may go there. The principle may seem to follow, but does not. For suppose you put a spot of green in the middle. Given that, it is impossible for you to array paint in such a way that it becomes false that there is a green spot in the middle. For the most elementary of logical truths - if P -> P – is not mysteriously suspended in the realm of painting. So you cannot get everything you want, if what you want is (in some addled cross between Macbeth and Hamlet) for a spot to be and not be.

This seems trivial – hardly worth mentioning - but all the difficulties of the art form flow from this logical point. We have all, in our time, set out to produce a resembling sketch of something. A few strokes in we realize, with dismay, that there is no way for this set of lines to be made to resemble that. Or, if we manage to clear this hurdle, we may get hung up on the next: we may realize that there is no way to make our sketch look like our subject and be aesthetically satisfactory. We wad up our botched work in disgust and start again; or take up some less arduous pastime than the production of fine visual art.

There are exceptions, but only ones that prove the rule. There are techniques that will produce the optical illusion that a dot of green in the middle is not green, or not in the middle. It is also obvious that representational resemblance, let alone aesthetic satisfaction, supervene on arrays of color in complex fashion. Picasso’s famous quip makes this point (in response to the complaint that his portrait of Gertrude Stein did not look like her): “Don’t worry. It will.” These exceptions only prove the rule, however, because techniques for making a central green spot look otherwise – let alone for engineering a time-lapse resemblance to Gertrude Stein – are highly particular. You cannot realize them just any old way, just by wanting them.

The thing that makes painting hard is not landing the color you select on the spot you select (as though the trick were analogous to hitting the bullseye with your dart.) Yes, you had better not tremble, but the hard part is reliably sensing and correctly anticipating what overall effects your strokes will have. You have to be able to sense and anticipate because you cannot will them. A green spot is a green spot and has all and only the effects a green spot will have, whether you like it or not

Getting back to our principle: it is perhaps true in a trivial (dart-throwing) sense that you can do anything you want in painting. But the deeper truth is that wanting is not enough, because usually what you want is not for this spot to go there; rather, you want the spot – together with all the others – to have specific, collective effects. You want them to sum to an attractive portrait, for example. This is not something you can directly will, however much you want it. Either all the bits of color add up to the desired thing or not. Usually not, hence all the bad paintings of the world. (It isn’t that artists don’t want well, for the most part; yet mostly they don’t paint wisely.)

A clear corollary - though it would be rather hard to prove: certain potentially desirable effects are simply impossible, i.e. unproduceable. I am presently going to adduce some comic-book examples of fictional worlds, for the worlds in which superheroes dwell are morally deviant. By way of warm up, here is a lament by one comic book writer, Mike Oeming, glad to have escaped the world of children’s book cover art, in which demands are made for sillier things than musclemen in tights:

“Nothing on this earth is worse than a book editor. They know everything about words, but nothing about pictures. Example, they say they want a Zebra, so I give them a Zebra, but then they ask for no stripes. I say, then it will look like a horse … No, they say, they want a Zebra, just no stripes. So I do it … Then they say it doesn’t look like a Zebra. This story had something to do with a donut. So I drew one in the background like they wanted. Well, I put sprinkles on it, because without sprinkles of something, it just looks like a tire. They say “make it glazed,” so I do and it looks like a tire. I assure them with colors, it will look like it’s glazed. Then they say it looks too literal, it should be a “cosmic donut”.”

Oeming eventually manages to make the donut look a bit like it is ‘giving birth to the universe’; but it satisfies no one. Of course, this might be put down to inadequate technique. Or one might confuse the issue further by drawing a stripeless horse-like creature and writing ‘Zebra without stripes’ underneath. Will this ‘look like’ a zebra without stripes or not? A trick question. To repeat: it is hard to prove that any visual effect at which one might conceivably aim cannot be produced by any means. I say no one can produce a somber and moving portrait of the Madonna and child that is also a humorous picture of dogs playing poker. But this may be just the temptation some fiendish draftman needs to go forth and contrive an astounding duck-rabbit of a work that is both, depending how one looks at it. Yet despite the lack of proof-procedures in this area, it seems plausible to the point of self-evidence that there are limits to the range of effects it is possible to produce by arranging paint in two dimensions. Ergo, wanting these effects is not enough. Ergo, painters cannot – even in principle - do everything they might want.

All of which is obvious, now that I mention it, since no one thinks painting is easy. I have belabored the obvious because something analogous, and a bit less obvious, is true of fiction. Here it is harder to formulate the point, hence perhaps harder to see it, because – quite frankly – it is not clear what fiction is made of.

Paintings are made of paint. It is tempting to say stories are made of words. It would be better to say stories are made of what their words mean (for it is possible to tell the same story in different words.) But now we pull and the thread just keeps coming. Such-and-such a string of words on the page may mean that Hamlet killed Polonius. But what Polonius’ death means is a further (non-semantic) question, yet essential to the story. His death means Hamlet is guilty of murder; that the arras has at least one hole; that Hamlet must dispose of the body if he does not want it to be found; that Ophelia is fatherless; that Gertrude is stunned and Laertes presently furious; that Hamlet will be packed off to England; that Rosencrantz and Guildenstern will die, so forth. Polonius’ death may also mean something like: deep existential crises are invariably attended with tragic accidents. Or: sometimes it can be strangely noble to stab an innocent man through a curtain. If it does mean anything of the sort – and a nice question that is – it is another nice question whether we should say that this meaning, such as it may be, is part of the story (let alone in the fictional world of Hamlet.) And I should have mentioned at the start that it is possible to tell stories without words. Pictures and pantomime may suffice (though the line between symbol and gesture and word may get a bit grey.) And it is common to mix words and pictures and other objects. The script for Hamlet is no more the story of Hamlet, one might say, than a cake recipe is a cake. Indeed, the script is a recipe for a recipe, i.e. contains directions for concocting a representation of a story, in which actual humans stand for non-existent humans. And, come to think of it, there is no conceptual link between fiction and narrative, though the two tend to go together. (Equating fiction with stories is a sort of bookstore-based pun on the distinction between fiction and non-fiction books, on the one hand; and fictional and non-fictional modes of representation, on the other.)

Confronting this analytic thicket, all that seems quite clear is that fiction is typically a complex, mixed-media production of sound and sense, psychological association and logical implication, so forth. It is, to repeat, hard to specify what stories are made of. Yet it seems clear, whatever the stuff may be, that it is no less essentially stubborn and tricky to work, in its way, than is the stuff of painting. In stories, in a sense, the author can do anything, just as you can put a spot of paint anywhere. But authorial actions have consequences that exceed, hence perhaps defeat, authorial desire. This point is made amusingly in Edward Gorey’s short illustrated work “The Unstrung Harp, or, Mr. Earbrass Writes a Novel”. The protagonist, Earbrass, is at one point exhilarated at the swift progress he is making on his latest work of fiction, The Unstrung Harp. Specifically, “he cannot help but feel that Lirp’s return and almost immediate impalement on the bottle-tree was one of his better ideas.” But awkward consequences obtrude:

“Mr. Earbrass has finished Chapter VII, and it is obvious that before plunging ahead himself he has got to decide where the plot is to go and what will happen to it on arrival. He is engaged in making diagrams of possible routes and destinations, and wishing he had not dealt so summarily with Lirp, who would have been useful for taking retributive measures at the end of Part Three. At the moment there is no other character capable of them.”

Here again the specific trouble can be put down to inadequate technique. Earbrass is a bad writer. But the general point is clear and valid. When you put something in your fictional world – e.g. a generous portion of bottle-tree through Lirp - its placement has effects. You cannot will those effects away (without going back and erasing whatever caused them) because the effects are, by and large, not direct functions of authorial will. You cannot, as Macbeth devoutly wishes he could, murder someone and then ‘trammel up the consequence’ – i.e. make it so the person is not, after all, perhaps inconveniently dead. Again, there are exceptions, but only ones that prove the rule: you can make your tale a supernatural one in which the dead live. Or an implausible one in which details are sloppily fudged. Or an absurdist exercise in frank and forthright logic-flouting. But these authorial prerogatives are not techniques for defeating consequences generally, only for finessing them in limited and highly specific ways. And any and all techniques for avoiding consequences generate unavoidable consequences. You cannot trammel them all up, and there are always more. “Oh, what a tangled web we weave,” so forth.

Thus, as in the painting case, in fiction most of the results one typically most wants cannot be directly willed. One wants to conjure a psychologically realistic protagonist who is troubled and antisocial yet sympathetic; or one simply wants to construct a locked-room mystery to which there is only one possible solution. But such things can only be achieved indirectly, i.e. implied by many small steps. (‘Show, don’t say’ is often invoked as a normative principle of fiction-making. In fact, it is often flatly descriptive.)

And so often what one gets, since one’s steps go astray, is not what one wants. It is no objection to this that some authors do not make missteps. Even if that were true, it would not follow that some authors can do whatever they want. Ideally, some authors are skillful enough to do nothing that has consequences inconsistent with what they want. But a person who lives such a well-arranged life that her desires are always fulfilled, never frustrated, is not, per se, omnipotent, merely very happily placed. The principle of poetic license alleges sheer authorial omnipotence, not happy placement, hence is false.

§4

This passion, and the death of a dear friend, would go near to make a man look sad. - Shakespeare, Midsummer Night’s Dream, (V.i)

Consider a possible objection to the foregoing. (It is not sound, but instructive.) You can simply stipulate anything and everything. No indirection necessary. Here, for example, is a locked-room mystery:

The Murderous Mystery of Thisby

'Pyramus and Lion were locked in a room with Thisby, who was alive when the lights went out but is now dead. Whodunnit? Pyramus couldn’t have. Hence this mystery has one possible solution. Lion was the murderer!'

The obvious response is that this mystery is not good. On the other hand, if it weren’t for bad mystery, some stories wouldn’t have mystery at all. The request was for a locked-room mystery. Period. But now, it seems, we need only narrow our specifications. Suppose, as seems likely, the author wants to write a good locked room mystery, in the classic, English style. Since good mysteries have to be painstakingly implied, not woodenly stipulated, it follows that authors cannot simply do what they want. They are not omnipotent. If they were, writing good locked room mystery novels would be trivially easy. The principle of poetic license is not supposed to deny this – no one would deny it – but, in effect, the principle denies it. The principle implies that if you wanted to write a good locked room mystery, and if you wanted to be able to generate this effect directly – i.e. without all the botheration of characterization, props, careful blocking, etc. - you should be able to do so.

But there is a twist. Suppose the objector now points out that the fact that a mystery is good is not a fact in the story. It is a fact about the story. The principle of poetic license, as formulated, states that authors can make any proposition true in fictional worlds. What is the counter-example, in this case?

Answer: something like, ‘X, Y and Z imply the truth of P’. This is not something you can directly will to be true in your fiction world, because – per itself – a part of it is to be made true not by you but by X, Y and Z. And this sort of case will actually be the norm. You, the author, will most likely want a certain something to follow not from your direct will but only from other things in the fictional world (e.g. that the story as a whole has the proper characteristics of a mystery of a highly specific sort.) Ergo, a whole class of authorial desires for propositional truth in fiction cannot be expressly stipulated true: namely, truths to the effect that a given proposition is true only because it follows from other fictional truths (hence not because the author wills it.)

But does it really follow that the principle of poetic license is false? It would seem all that necessarily follows is that some propositions have to be made true indirectly. But perhaps there are no propositions that cannot be implied by some cunning, authorial trick. Ergo, perhaps the principle is true. This is already a half-step back from the position that the principle is true. And it strikes me a full step back is almost certainly necessary. Why should authors be able to imply anything they like? As per above, authors cannot trammel up all consequences of everything they do – no more so than the rest of us can, in the real world. Mr. Earbrass wants to impale Lirp on the bottle-tree, then have Lirp take retribution, and not to have anything illogical or supernatural happen in his story (let us suppose.) It seems he is stuck. He wants a compound proposition, (P & Q) -> - (R v S), to be true in his fictional world. His poetic license expires before he gets this far out.

There are in fact further complications that could be explored at this juncture, having to do with the fact that it is often not clear whether a given truth is true in the story, or merely about the story. Mostly these mixed cases have to do with literary conventions, genre rules, so forth. Shakespeare’s characters speak poetry; characters in opera and musicals sing songs. Yet it is not the case that the characters are supposed to be speaking poetry or singing. James Bond is the hero, yet it is not a fact in the story that he is the hero (although it is true in the story that he is a hero.) And it is not a fact in the story that all Bond’s missions exhibit truly bizarre similarities. (If it were a fact, Bond would notice, for he is very observant. On the other hand, all Bond’s missions are bizarrely similar.) In parodies of Shakespeare and opera and genre fiction generally, an inevitable joke is to have the characters suddenly notice the absurdity of the rules they are following. That is, in parody worlds (unlike the worlds they parody) rules of genre are true in the world.

Without delving further into these intricacies, let me briefly state what all of this has to do with the alleged problem of imaginative resistance to morally deviant fiction. First, there is a problem in trying to work out whether the moral of a story is a moral in the story. Resisting the moral of a story is refusing to accept false claims about our world: nothing whatsoever to do with imaginative resistance to fictional moral deviance, then. (This is close to Gendler’s point in her paper, hence the reason why I think she should say there is no problem of imaginative resistance, according to her.)

Relatedly, there is a problem – obscured, I think, by the vagueness of ‘imaginative resistance’ – deciding whether the difficulty with producing morally deviant fictional worlds is supposed to be a difficulty achieving literary success in the genre. Or whether it is supposed to be a difficulty producing any example of the genre whatsoever.

Gendler and Walton – and others: Brian Weatherson, for example - would say they mean the latter. But it is not at all clear to me this is consistent with their procedures. Gendler, in particular, takes the literary success of her fable, “The Tower of Goldbach”, as proof that it does not provoke imaginative resistance. But why should it not be aesthetically satisfactory to throw oneself into a story, only to be bounced out, imaginatively speaking? Why should something worthy of the name ‘imaginative resistance’ not be a source of literary success? (Isn’t this Lewis Carroll’s best trick, most days? Isn’t reading complete nonsense fun? It does not follow that it isn’t complete nonsense, surely.) If this is right, literary success hardly proves absence of imaginative resistance.

Conversely, Walton, Gendler and Weatherson all offer little narrative bits and pieces – sentences, or short paragraphs, containing morally deviant oddities – and treat these as prima facie proofs that morally deviant fictional worlds cannot be produced. But would it not be more plausible to regard these as evidence that stories about morally deviant worlds tend to be aesthetic failures? Here, for example, is “Pie”, a short-short tale by Brian Weatherson:

“It had been a quiet day at the APA Pacific until the young guy talking about imagination livened up proceedings by throwing a custard pie in George's face. George seemed momentarily discomforted by the surprise attack, but in the audience most people were laughing. Some of them were doing that nervous "What will he do next" laugh, but most were really amused. Some Kantians started tut-tutting about the use of poor George as a means not an end, but they were wrong to complain. Everyone else was made so happy by the funny surprise, that clearly throwing the pie was the morally right thing to do.”

In another paper, “My Favorite Puzzle”, Weatherson distinguishes between alethic puzzles regarding stories like this – i.e. puzzles about whether sentences like the final sentence of “Pie” can be true – and aesthetic puzzles – i.e. puzzles about why we don’t think offerings like “Pie” are great literary achievements. Weatherson says he sets the aesthetic puzzle aside. Likewise, Gendler hastily brushes aside ‘niceties of literary interpretation’ at one point in her paper. But it is far from clear to me that our reaction to “Pie” (and to similar examples concocted by Gendler and Walton) are anything but literary reactions, i.e. aesthetic ones. I rather suspect that literary niceties are close to being alpha and omega hereabouts. We do not see the sense of stipulating that it was right to throw the pie. By contrast, Gendler’s fable has a familiar literary form. It can be coordinated with familiar genres. It has a clever ending. There is a point to writing a tale that has a witty twist in the tail. So long as we can cling to the point, we are happy. Then we don’t mind that we are completely imaginatively resistant to the notion that 7 + 5 could fail to sum to 12.

Speaking of points, let me come to one regarding morally deviant fictional worlds. It seems to me that it might well be (for all anyone has shown) that one can stipulate any old thing one pleases – murder is good, saintliness bad – and thereby quite easily produce an instance of the genre of morally deviant fiction. This achievement will almost certainly be of little or no intrinsic literary or aesthetic interest. (Just as ‘the murderous mystery of Thisby’ is not a successful entry into the locked room mystery field.) The trouble is not that it is offensive – who could be scandalized by “Pie”? - but that it is not going to make it into literary anthologies any time soon. And this is nothing new under the sun. Most fiction is really not very good, begging Weatherson’s pardon. (Another word for imaginative resistance might be ‘boredom’, perhaps. But if that turns out to be the case, the whole investigation has become hopelessly ill-defined.)

What would be interesting, by contrast, would be implying – rather than baldly stipulating - the truth of some deviant morality. How could one set about implying, rather than directly stipulating, that morality itself is different on some fictional world than it is in ours? It is easy to see why this would be hard. How do you work your way into a deviant morality without simply stipulating something deviant about morality? In fact, I think authors manage to do it all the time, even in quite humble contexts; and this is a quite interesting fact about the craft of fiction writing.

Let me conclude this section with an analogy I hope will guide us in what follows.

Mike Oeming is not the only artist to find his patron’s demands disagreeably demanding – cosmic donuts and stripeless zebras and all. One of Giovan Lorenzo Bernini’s contemporaries wrote of the great sculptor:

“He no longer wishes to make sculpture portraits after paintings, because it is tedious and difficult. He spent fourteen months working on the portrait of Francesco d’Este.”

Why is it so difficult to make sculpture portraits? Another acquaintance of Bernini explains:

“Talking about sculpture and the difficulties encountered in order to succeed, especially in achieving a resemblance in marble portraits, he told me something remarkable and then repeated it on many occasions; if someone were to dye his hair, beard and eyebrows white and, if it were possible, the pupils of his eyes and his lips too, if he then showed himself in this state even to people who saw him every day, they would barely recognize him. Therefore it is very difficult to achieve a resemblance in a marble portrait, which is all one color.”

In a sense, this is unsurprising (maybe you knew it already.) When you draw figures in the distance, you make them smaller on the canvas. You have to trick the eye. Why should sculpture be different? On the other hand, it does make you look at marble sculpture differently when you realize for the first time that it isn’t actually shaped at all right (otherwise it wouldn’t look right.) And it makes sense that portraits, where precise resemblance to an existing human model is so important, would be particularly difficult. (We humans are so good and careful when it comes to recognizing faces. Any little thing wrong will look really wrong. Yet the shape has to be quite wrong to look right.) And it would no doubt be extra difficult to have to transform one transformation – i.e. a projection onto a two-dimensional field – into another – i.e. a pure white bust, without losing the resemblance.

Now, my metaphor: morally deviant fictional worlds are (at least in the most interesting cases) a bit like carving a portrait bust in marble. You might think the difficulty is due to the fact that morality is hard stuff. And you wouldn’t be wrong. (You shouldn’t murder people – really, you shouldn’t.) It’s also the case that the stuff is frustratingly soft. (This is what has vexed some of Walton’s readers - Michael Tanner, for example. Walton may seem a veritable babe in the wolds of moral complexity, relativity and pluraity. And there’s something to that, but not as much as some of Walton’s critics may think. And we will get to that.)

But the real trick ends up being: you carve your morally deviant fiction wrongly – just right - so your audience approves the result because it looks right, even though it’s actually wrong. It’s a neat optical illusion. And now having secured your audience’s tacit consent to a rather alien shape, you are able to imply deviant things (I suppose the sculptural analogy would be: you paint the surface of the bust that everyone has already agrees looks just like the man - and what strange results you will achieve.) The audience may find itself nodding to deviant results, and never be quite able to put its finger on the point at which the departure was made. (Not that it's so well hidden, but you do have to think about it.)

And it’s rather interesting to lay out how all this is done. I am most interested in the ways in which you can use conventions of genre to pull off the trick. But first let me make a general survey of all the varieties of fictions - there are many - that seem to me thoroughly morally deviant.

... So that's where it leaves off, for now. What follows has been indicated rather impressionistically in my previous post. I want to catalogue (to mangle James) the varieties of morally deviant fictional experience. There are many. The specific variety that interests me most is a function to genre - varieties of heroic epic and pastoral (including mock pastoral). All this follows Empson. I want to talk about how authors use various conventional techniques to dampen down moral implications, artificially; then others take the same forms and - as it were - release all that coiled up moral energy in mocking or affectionate springs. I want to talk about Homer, and Shakespeare's treatment of Homer in Troilus and Cressida; and a bit about Midsummer Night's Dream - the rude mechanicals and their play. Do you stage it so that the aristos are dabbing tears at the end , despite themselves, or not? I want to talk about how, oddly, many of the same techniques are employed in comic books. The insane, deviant moralities of superhero comics are harnessed and released to considered effect by the likes of Alan Moore, Kurt Busiek, Bendis & Oeming. They do exactly the same things to the men in tights that Shakespeare does to Homer, oddly. I want to talk about the trick in "Unbreakable" (whether you think it works or not), and about Charlie Kaufman's "Adaptation", in which the man makes fun of himself for writing a screenplay to "Through A Scanner Darkly". (There is a joke in "Adaptation" that I'll bet very few people have gotten, but I think I have.) In a weird sort of way, it's exactly the trick that the marble bust metaphor suggests. A distorted form is let in, because it looks right. Why does it look right? Because genre tricks your eye. Then you give the distortions free rein.

Of course, I won't actually do all this. That's why I'm mentioning it here. We can always dream of cramming it all in, just the way it rolls around in our heads. Part of me thinks I actually have much more interesting things to say about the similarities between Shakespeare and comic books, seen through an Empsonian pastoral lens. The magic of genre fiction. (It isn't just comic books, though there's nothing wrong with comic books. We live in a genre-saturated world, and there are odd things to be made of it.) I oughta just write about that and leave Gendler and Walton clean out, maybe. Analytic philosophy and cultural studies sit oddly together. I always have this problem. That's why I don't publish.

Comments

[Ray]

This is specifically not the sort of thing you were looking for, and I'm specifically not the sort of person you wanted to respond, and it fits your earlier post on the subject better than this one, but having seen this red cloth waved twice, I cansk stands no more.

There *is* such a thing as "moral [speculative] fiction," and it's *called* "science fiction." Thinking that science fiction only speculates about physical sciences is like thinking that academic journals only publish accounts of university life. I know you read Gene Wolfe, and he portrayed a professional torturer as Messiah, for chrissake.

"1) In killing her baby Giselda did the right thing; after all, it was a girl."

My own first guesses as to the context of that sentence would be:

a) This is a sf/fantasy story set in a culture that practices female infanticide. (First lines of that sort are by no means rare in Asimov's or F&SF.)

b) This story is freely translated from a culture that practices female infanticide. (I've heard such cultures have existed.)

One of the things that makes literary reading *literary*, as opposed to a survey of fools and crazies whose garbage laid the groundwork for our status at the pinnacle of creation but who should now be hygienically discarded, is our attempt to deal with the alien artifact as something worthwhile despite its blatant out-of-placeness.

Science fiction replaces that "despite" with a "because of."

And if a horror or a suspense story began that way (if, say, I knew the author was Patricia Highsmith), would we be reading it *because* the narrator is "mad, or bad" -- as a useful reminder to us not to have children with mad or bad people -- or would we be reading it because we take pleasure in imagining a different version of what's mad or bad?

Walton and Gendler have locked themselves into an impoverished closet-sized room and then imagined that their starvation constitutes a mystery. "Pie" reads as trivial, but "Love Is the Plan, the Plan Is Death" does not. Neither do the stories of Ted Chiang, which resemble the "thought experiments" of these essays and comments (changing the laws of math, direct intervention by God) except in having attained the state of viable fiction.

[jholbo]

Ray, I mostly agree with you. I'm actually going to address this point directly a couple sections on, after offering all my counter-examples. The general form of the anti-Walton/Gendler complaint will be: Walton and Gendler cite Hume as the source of their problem, originally, and Hume is preposterously antique, in a (let's say, though this is just a start) pre-Romantic way. His ideas about what you can (and should) do with fiction are hopelessly narrow. And if Walton and Gendler are following Hume, then somehow they've managed to miss the 19th Century, not to mention the 20th. I think this is basically right. But there is a significant point to be made on the other side. In many cases - say, Ted Chiang's "Hell is the Absence of God" - the potency of the speculation is not due to the fact that Chiang has changed the moral rules. The potency is due to the fact that he has changed a lot else yet (we the readers feel, without really needing to be told) the moral rules stay the same. The the difficulty of conceiving morally deviant worlds becomes a structural tool employable by speculative fiction writers. They can mess with everything else, and count on something staying the same.

A lot of cases you cite are not actually cases of deviant morality, as W & G understand them. For example, you offer two readings of the Giselda line, neither one of which imagines changing the world to make morality different - just changing people's BELIEFS, to make their view of the world differently. W & G grant that. Of course, a very sensible theory of the whole business is that morality is a function of beliefs, so if you believe different things, morality changes. Well, we'll get on to that. The short version of it: what separates you from, say, Hume, may be that you are an incredibly tolerant pluralist about morality - especially in fiction - whereas Hume was a narrow stick-in-the-mud. (A bit funny to say this about the man who said it would not be irrational to prefer the destruction of the world to the scratching of his little finger. But the fact is he didn't. And he is a bit narrow in his approach to literature.) Moving right along: but if your view is that such-and-such a plural view is valid, then presumably your view is that it is valid in OUR world, hence imagining a world in which some subset of is this plurality is the rule does not count as imagining a DEVIANT world. The more tolerant you are, then, the farther you have to go to find anything that counts as actual deviance. So the proposition that deviance is hard to produce is supported from a different direction. Absolutists don't allow the stuff, and relativists can't produce it, by default.

The thing I mean to discuss in the essay - and your negative reaction is a case in point - is that we tend to think of morality as more plastic, in fiction, that it really is. There's a kernel of insight to Walton's point of view, though the view is indeed wrong - so the kernel just provokes good questions at most. In a lot of cases, it's the very hardness of morality, even in the face of poetic license, that allows authors to do their weird speculative tricks.

[T. Gracchus]

How do we disentangle deviant morality from deviant political justice? We all read happily stories which presume the justice of hereditary rulers and dramatic class differentiation. What are they examples of? Second, to repeat an example raised earlier, Sade does conceive of a morally deviant world and, while the writing has almost every other defect, it is not hard to understand what he proposes. Third, aren't large parts of popular cinema made up of examples of morally deviant worlds? The Unforgiven is built, in large part, on the premise that an appropriate punishment for disfigurement is assassination. Or consider the Tom Riply movie out last year. Then again, maybe I am just missing the whole point.

[David]

Though you can't put this into a journal article due to its terse nature, it may be far easier to simply acknowledge that we cannot transcend our human condition. The best fiction writers are those who are able to appeal to just that humanity which we all possess, rather than a certain particularity of time and place. This explains why we can be either conflicted or relieved or in broad agreement with the end of "Cousin Bette;" this is possible though we don't live in early 19th century France. It also explains how lesser writers fail because they cannot create a world which is sufficiently compelling to humans ie. it doesn't resemble our human world enough to interest us. This would apply to writers trying to tell one-dimensional historical tales or to science fiction scribes trying to get us to embrace stories where infanticide is a common, acceptible practice. One need not agree with a writer's characters but the characters and their world have to be more than just a cardboard cut-out; this is only possible if their world bears a striking--though inexact--resemblance to ours.

[TomD]

Surely "Crime and Punishment" is a refutation of the thesis that morally deviant fiction meets insuperable imaginative resistance.

Indeed, any fiction with a villain whose interior psychology is convincingly depicted is a candidate refutation. We begin to get under the skin of said villain, and inevitably begin to cast him as partly a hero, because - given the emotional charge that the author invests in him - his actions feel justified.

In most examples, we move outside the villain's emotional orbit at some point in order to pass judgement; but at any point where we were completely in sympathy with a morally deviant character, we were experiencing a morally deviant fictional world. This can happen virtually without any imaginative resistance, because some morally deviant things are just so satisfying to imagine oneself doing, and we respond gratefully to the invitation to turn off our moral censor.

The ancient Greeks understood it, if we take catharsis to be the moment at which we allow ourselves to be emotionally convinced of the rightness of a morally deviant act. Precisely because the action is fictional, we allow ourselves to live temporarily in a world where (say) revenge killing is right.

Revenge is a prime candidate: even though most of us might agree that revenge is wrong, we root for the character who's out for it. Within the story, revenge is right. And this is often true just *because* it is a story: because, for example, the story ends at the point of the "hero"'s triumph and any further complications (which would make it clear that revenge was wrong) are glossed over.

The Screwtape Letters, Othello, and Otello the opera, are unsatisfactory precisely because they don't present a psychologically satisfying villain, but merely an attempt at one: their authors failed to sympathize fully with the villains.