What follows is a guest post by Amy Kind. Amy is Professor of Philosophy at Claremont McKenna College. Having received an AB summa cum laude from Amherst College, Professor Kind received her PhD in philosophy from UCLA in 1997. Although she has broad interests in the philosophy of mind, most of her research centers on issues relating to phenomenal consciousness and issues relating to the imagination. Her work has appeared in journals such as Philosophy and Phenomenological Research, Philosophical Studies, Australasian Journal of Philosophy, and The Philosophical Quarterly. She is currently at work on several edited collections, including The Routledge Handbook of the Philosophy of Imagination and Knowledge Through Imagination (co-edited with Peter Kung), which is under contract with Oxford University Press.
In recent decades, imagination has proved enormously important in philosophical discussions of aesthetics. Not only does imagination seem to play a crucial role in the creation of works of art and literature, but it also seems to play a crucial role in our engagement with such works. But what is imagination? This question has proved enormously difficult to answer, and there are obviously many aspects to addressing it adequately. Here I aim to focus on only one such aspect and only in a very preliminary way. When we engage in an imaginative exercise, our imagining has a certain target – perhaps an object, or a proposition, or a state of affairs. So the question motivating this post can be put as follows: When someone aims to imagine some target T, what makes it the case that she succeeds in doing so? In what follows, I hope to move us closer to an answer to this question by fleshing out our intuitions on some cases.
As my point of departure, I’m going to consider Olaf, the dimwitted but lovable snowman from Disney’s recent movie “Frozen.” Olaf is brought to life by Elsa, a woman blessed (or cursed) with the ability to turn things to snow or ice. Unable to control her magical powers, Elsa accidentally unleashes an eternal winter on the kingdom. Though Olaf hasn’t had much experience with heat, that does not stop him from dreaming of summer; as he tells us, “sometimes I like to close my eyes and imagine what it’ll be like when summer does come.” His imaginings are whimsically depicted in the musical number, “In Summer” (and I highly recommend taking the two minutes to watch this clip of it):
Presumably Olaf would be quite disappointed if he ever were to discover the truth about what happens to solid water when it gets warm. He’s not going to be able to run through a field of flowers in the midday sun, or frolic on the beach, or enjoy a refreshing drink while soaking in a hot tub. He may be a magical snowman, but he’s still a snowman. So any experience that Olaf could possibly have of summer would have to be a very short one, for there’s no getting around the simple, awful truth he’s yet to learn: Snow melts.
Clearly Olaf’s imaginings have in some way gone astray. But how, exactly, did they go wrong? There seem to be two ways we might diagnose his mistake:
(1) Though Olaf has indeed imagined his target, he’s done so very badly. He’s mistaken about what would happen to a snowman in these situations.
(2) Olaf hasn’t really imagined his target. Though he aims to imagine a snowman in summer, and in the sun, and on a hot beach, he hasn’t succeeded in doing so. Rather he’s imagined something else – perhaps he’s imagined a snowman in a season that occupies the summer months but doesn’t have the usual characteristics of summer, or perhaps he’s imagined a creature who superficially resembles a snowman but who’s not made of snow, or so on.
The first of these descriptions seems to me to be the more natural one, at least pre-theoretically. Surely that’s how Olaf himself would describe the situation. As Olaf lies melting after having been exposed to the sun, it seems to me that his regretful musings would be much more likely to be something along the lines of: “Oh summer heat, you were nothing like I imagined you” than to be something along the lines of: “Oh summer heat, I never really managed to imagine you at all.”
Based on philosophical discussions of modality, however, it looks like many philosophers are inclined toward the second of these descriptions. To give just one example, consider Saul Kripke’s famous discussion of a wooden table:
Though it seems to us that we imagine thistable made of ice, we are mistaken – really, we imagine something else. For this table could not have been made of ice. Likewise, then, though it seems to Olaf that he imagines snow surviving exposure to heat, he must be mistaken – really, he is imagining something else. For snow could not survive exposure to heat.
The temptation to opt for descriptions of type (2) in cases like Olaf’s is perfectly understandable; we want to be able to learn from our imaginings, and we wouldn’t be able to learn from our imaginings if they were unreliable in the way that descriptions of type (1) suggest. But I think this temptation should be avoided. We can’t wish away the unreliability of the imagination simply because we hope to put it to certain philosophical use. And once we set aside these external motivations, there seems to be very little reason to prefer descriptions of type (2). To my mind, such descriptions are motivated only by antecedent intuitions that certain imaginative targets are impossible, combined with the claim that we cannot imagine the impossible. But whether we can imagine the impossible is precisely what’s at issue.
Moreover, consideration of common imaginative projects helps to support the naturalness of descriptions of type (1) in a variety of cases. “Motherhood is nothing like I imagined it,” an exhausted new parent might snap. “Obama’s presidency is nothing like I imagined it,” a formerly optimistic liberal might complain in frustration. Reality turned out to be quite different from what these imaginers had predicted via their imaginings, but that doesn’t give us cause to deny that the exhausted mother had really imagined parenthood, or that the frustrated liberal had really imagined Obama’s presidency. It’s not that their imaginings missed their imagined targets, but that they were mistaken about certain salient properties of those targets. Sure, we might sometimes say things like, “No one can really imagine parenthood unless they’ve been through it,” but in my view, what we’re really saying in such cases is that “No one can really imagine parenthood correctly unless they’ve been through it.”
The plausibility of type (1) descriptions is also supported by examples of a different sort. Consider first a passage from Wittgenstein:
Someone says, he imagines King’s College on fire. We ask him, “How do you know that it’s King’s College you imagine on fire? Couldn’t it be a different building, very much like it? In fact, is your imagination so absolutely exact that there might be a dozen buildings whose representation your image could be?”—And still you say: “There’s no doubt I imagine King’s College and no other building.” (Wittgenstein, The Blue and Brown Books, p. 39)
Or consider the common experience of imagining fictional characters. When reading about Harry Potter’s first encounter with a Dementor aboard the Hogwarts Express (in Harry Potter and the Prisoner of Azkaban), someone might naturally imagine the Dementors. Let’s suppose, however, that the face of the creature that the reader has mentally conjured looks remarkably like a deep sea hatchetfish:
But let’s also suppose that the reader has never heard of, nor seen pictures of, deep sea hatchetfish. That her imagining resembles them is purely coincidental. Would we describe her as having imagined a sea creature instead of a Dementor? To my mind, the answer is clearly no. The accidental resemblance between what she imagined and the hatchetfish does not make it the case that she failed to imagine a Dementor, or that she succeeded in imagining a hatchetfish. And this would be the case even if hatchetfish didn’t look very much like Dementors.
(But would this be the case no matter what her imagining was like? Suppose the image conjured resembles a fluffy bunny rabbit. Here I’d admit that I’m considerably less inclined to agree that she’s succeeded in imagining a Dementor – after all, they’re described as “glistening, grayish, slimy-looking, and scabbed, like somethng dead that had decayed in water.” So I don’t want to rule out the possibility that some imaginings are so bad that they miss their target altogether. The question, then, is where to draw the line.)
With these cases before us, then, I’d suggest that we’ve started to make progress on an answer to our initial question. In particular, I think that the cases suggest that an imaginer’s intention in pursuing an imaginative project – her intention to imagine a particular target – goes a long way towards latching her on to that target.
This seems to me to have important implications for philosophical claims about what we can and can’t imagine. For example, it starts to seem to me considerably less plausible that we can’t succeed in imagining what’s it’s like to be a bat. Likewise, why shouldn’t we admit that we can imagine philosophical zombies? Perhaps these things are imagined quite badly, perhaps as badly as Olaf imagines summer, but that is quite different from claiming that they can’t be imagined at all. Of course, this also means that we would have to reexamine claims about whether imagination can serve as a reliable guide to possibility. But it strikes me that we need to pay considerably more attention to what we can or cannot imagine before we can even begin to make progress on the question of the role that imagination has to play in the epistemology of modality.