Antonia Peacocke is an Assistant Professor of philosophy at Stanford University. She works on mental action and aesthetics.
A post by Antonia Peacocke
In imagining something, you can bring to mind mental images all of whose imaged properties—including colors, shapes, smells, sights, sounds, and other perceptible and sensible qualities—underdetermine the content of what you imagine overall, or its total (imaginative) content. You can take a mental image of a horse that doesn’t itself ‘picture’ any motion whatsoever, and use it to imagine something that necessarily involves motion, like a horse galloping down a track. You can use a mental image of an apple to imagine a perfect hyperrealist sculpture of an apple. You can use an image of a man you can picture in order to imagine James K. Polk, whose precise features you just don’t know. Let’s call cases like these “underdetermination cases.” I’m going to set out a couple of views that aim to understand the content of imagination in these cases, and then raise an unsolved problem for such views.
There may be a type of imagination (so-called ‘propositional imagination’ or ‘imagining that’) which does not involve mental imagery at all. But the cases of imagination I’ve just described aren’t like that; they all involve images, just not in a way that fully determines the content of those episodes of imagining. Following Kind (2001), I think it’s a mistake to respond to these cases by denying what she calls “the essentialist claim:” the claim that images play an essential role in this kind of imagination. Mental images are certainly doing some important work here, and what’s more, they are at least partly settling the total content of these imaginings.
To see why they play some content-settling role, consider cases of extreme mismatch between your mental images and what you are trying to imagine. If you only had an image (as) of a chimera when you tried to imagine a game of basketball, you just wouldn’t count as imagining a game of basketball at all. That would not be the content of what you imagine. You might have been thinking of a game of basketball while bringing to mind this mythical creature, but you certainly weren’t imagining a game of basketball as such.
So we shouldn’t respond to underdetermination cases by giving up on the idea that images play an essential role in imagination of the relevant kind, nor even by giving up on the idea that images at least partially determine the total content of an episode of imagination.
What, then, should we say about underdetermination cases? Following work by Chris Peacocke (1985), Peter Kung (2010) endorses a theory of the imagination that has two separable aspects. There’s the basic qualitative content of your imaginative episodes—the phenomenology of perceiving things and their perceptible properties—and then there’s assigned content, which adds yet more to what you perceive in the mind’s eye in this way. Some of what’s added are conceptual tags, or ‘labels:’ for example, a fluffy tan and white creature imaged in the mind’s eye might be labeled as a Corgi. These labels let you think about things under the categories to which they belong. The rest of assigned content is stipulative content, which is just as it sounds: you can stipulate, of the Corgi, that he is named “Fergus,” and belongs to Queen Elizabeth II, without modifying in any way the basic qualitative content of your mental image of this pup. This stipulation is something you do, and do with an amazing amount of freedom:
We have tremendous power and flexibility in imagining because we can fix via assignment what is the case in our imagined situation to an almost arbitrary level of detail.
In my terms, Kung’s model takes the total content of an imagining to be determined not only by basic qualitative content but also by the content you choose to assign to it.
Nick Wiltsher (2016) calls Kung’s view “the additive model” of the imagination. Accepting it is perfectly consistent with accepting Kind’s essentialist claim about the imagination, since, as Wiltsher points out, assigned contents are about what is imaged, and so plausibly require images in that capacity. But Wiltsher rejects the additive model as “otiose,” for the following reason: once we’ve said how the mental images in question are generated in the first place, the stipulations will be entirely unnecessary to explain any aspect of content.
Wiltsher claims that “you actively generate an image by deploying a concept,” and that concept use itself adds to content whatever else is present over and above the basic qualitative content of a mental image itself. He agrees with Kung on the point of freedom: the contents that extend beyond the basic imaged properties of an object are “in the gift of the subject.” But he takes himself to deny a key aspect of Kung’s view: since use of a concept is always needed “to generate even the simplest of object-images,” in a way that fills out the total content of the imaginative episode, there’s no work for any assigned content to do here. Where Kung thinks we need another kind of content to bolster the imaged content into the total content of an imaginative episode, Wiltsher thinks the generation of mental images ensures that the images themselves have a much richer content than Kung (or Peacocke) ever allowed.
I’m not so sure that Wiltsher’s proposal is really so far from the core of Kung’s theory. Where Wiltsher endorses a “unity view,” claiming that the additive model engages in “double-counting” on “imaginative achievements in the same experience,” Kung anticipated a version of this complaint, and disavowed one natural reading of the word “assignment”:
it incorrectly suggests that imagining is a two-stage affair, where we first conjure up some qualitative mental “picture” and then label or stipulate various things in or about that “picture.” Imagination does not work this way … The imagery comes with everything already labeled and stipulated.
Determining whether Wiltsher’s view is more than a notational variant of Kung’s will require much more of a detour into theories about the nature of imagistic content, and I won’t take that detour now. What I will do is raise a problem for both views.
In their own ways, each of Wiltsher and Kung take the non-qualitative part of imaginative content to be generated by your own action: it is “in the gift of the subject,” an expression of “tremendous power and flexibility in imagining.” There are many good reasons to take some imagination to be a matter of mental action, and indeed to take its content to be partially determined by what you are trying to do. But this cannot be the whole story about the content of imagination that extends over and above sensible and perceptible features imaged in the mind’s ‘eye.’
Here are two reasons why.
First, consider what happens in dreams during sleep. We often have dreams that are filled with vivid mental imagery—primarily of sights and sounds, but sometimes of even more. These images themselves come entirely unbidden; it doesn’t seem that there is anything I do that brings them to mind. What’s more, there’s nothing I can do to modify them or change my ongoing experience, as I usually can in waking episodes of imagination. (In his 2000 book Consciousness and the World, Brian O’Shaughnessy’s even suggested that dreaming consciousness involves a total “will freeze,” in which you can’t do anything mentally at all. For the time being, let’s set aside any thoughts about so-called “lucid” dreaming.)
This might be all well for Kung and Wiltsher, except for the remarkable fact that something like assigned content is just as obviously present in dreams as it is in waking life. You can ‘see’ the farmhouse from Ratatouille in a dream but take it to be your childhood home (even if, in fact, you grew up in a big city). You can ‘see’ the vast blue expanse of sky above you and take it to be an illuminated dome over a soundstage on which you live. Neither of these further contents derives just from the sensible or perceptible qualities in the image; these are genuine underdetermination cases. But there are no intentions that form the images in question, and no need for any concept deployment to generate such images either. Something in your brain puts on the light show at night, with no effort needed on your part. Insofar as Wiltsher’s or Kung’s view explains the most sophisticated content of imaginative episodes in terms of something you actively put there, neither one will be able to explain what goes on in dreams.
Second, consider the importance of imagination in inspiration. The language of inspiration is passive: we say “I was inspired,” using the passive voice. Inspiration strikes you, or it comes upon you. Inspiration is often indispensable in producing creative work—and not only purely sensible art like music or painting, but also narrative work like novels. What can come to mind in inspiration is not just limited to the kind of basic qualitative content Kung claims that mental images have. You might be struck not only by the mere looks of an eleven-year-old girl, but also (and simultaneously) by the idea that she’s a spy named “Harriet,” despite there being nothing in her appearance that gives that away (cf. Fitzhugh, 1964). Or you might suddenly find yourself not only gifted with a mental image of a four-sided equilateral polygon, but in the midst of a full-blown imagining of a square who is also a gentleman who yearns to travel (Abbott, 1884).
There is nothing particularly unusual about the suggestion that authors can be so inspired, in a way that brings to mind much more than sensible or perceptible qualities of things. But this, too, creates a problem for any view on which such further content is only ever granted to an imagining by prior conceptualization of the object of imagination—let alone an intention to imagine that kind of thing. If we allow the kind of richness into purely inactive inspiration that Kung only allows into assigned content, as I think we must, we cannot have a picture on which such content comes only from something you yourself do.
If we accept what I’ve said about dreams and inspiration as forms of imagination, we need not immediately give up on the nice proposals made by Kung and Wiltsher. We might just allow that they tell us about the total content of waking, active episodes of imagination. But we might want a more unified story of dreams, inspiration, and active imagination. We might be more struck by the similarities between these mental phenomena, and their rich contents, than their differences. If so, we should look for a new theory, one that can explain the total imaginative content of all underdetermination cases in precisely the same way.
References
Abbott, Edwin A. (1884). Flatland: A Romance of Many Dimensions. London: Seeley & Co.
Fitzhugh, Louise (1964). Harriet the Spy. New York: Harper & Row.
Kind, Amy (2001). “Putting the image back in imagination.” Philosophy and Phenomenological Research 62.1: 85-109.
Kung, Peter (2010). “Imagining as a guide to possibility.” Philosophy and Phenomenological Research 81.3: 620-663.
O’Shaughnessy, Brian (2000). Consciousness and the World. Oxford: Oxford University Press.
Peacocke, Christopher (1985). “Imagination, experience, and possibility: A Berkeleian view defended.” In J. Foster and H. Robinson, eds., Essays on Berkeley (Oxford: Clarendon Press).
Wiltsher, Nick (2016). “Against the additive view of imagination.” Australasian Journal of Philosophy 94.2: 266-282.