Recently, I have been reading work on the imagination model of dreaming by Colin McGinn, Ernest Sosa, and Jonathan Ichikawa. (This bears on one of the arguments I run in Chapter 5 of my dissertation.) This work is set-up as being in opposition to what has been called the orthodox or received view of dreaming. Thus, Ichikawa opens a recent article entitled “Skepticism and the Imagination Model of Dreaming” by writing:
According to the orthodox theory of dreaming, when dreaming, a subject has misleading sensations, which typically lead to false beliefs. Descartes assumes the orthodox theory when he introduces dream skepticism….
The orthodox theory can be contrasted with an imagination model of dreaming, according to which when I dream that p, I do not have misleading sensory experiences as of p and falsely believe that p: instead, I have sensory imagery as if p were the case, and propositionally image that p. That is, I simulate the experience of p. On this model, the experience of dreaming becomes more like the experience of fiction, or the experience of a vivid daydream. I do not falsely believe myself to be flying, and to have misleading visual percepts as of the tops of clouds; instead, I am imagining myself to be flying, and calling forth visual images of cloud-tops. (2008, 519-520)
I wonder whether the “orthodox view” is really so orthodox. No doubt dreaming is typically referred to in perceptual terms; and, given that, I can see why the imagination model might be assumed to be in opposition to the orthodox view. Nonetheless, it strikes me that what is at issue in much of the talk about dreams that is supposed to be perceptual, are in fact claims that are also made by advocates of the imagination model: I suspect that the perceptual talk often aims to express the belief that in dreams we can be acquainted with qualities such as colors and sounds that are essentially indistinguishable from those that we are acquainted with in ordinary waking perception. As such, the point is not so much to classify dream experience as being perceptual, as it is to claim a similarity between episodes of acquaintance in dreaming and waking perception. Thus, I hold that much perceptual talk about dreaming does not actually take a side with regard to the contrast between “sensory experiences” and “sensory imagery.”
My impression is that this is supported by what I have read of other philosophers making use of the example of dreaming; it seems that the focus is typically on the supposed phenomenological kinship between dreaming and waking experience. To be sure of this, and to compellingly make the case that this is correct, would require a systematic look at how dreaming has been used in philosophy since roughly Descartes. As that isn’t going to happen any time soon, let me just begin with a suggestive look at Descartes. In the Principles of Philosophy (1644) Descartes begins Part I with doubt and this includes raising doubts about perceptible things. Using the Miller and Miller translation, in Article 4 Descartes writes:
4. Why we can doubt perceptible things.
And since we are now only concerned with seeking the truth, we shall begin by doubting whether any perceptible or imaginable things exist: first because we perceive that our senses sometimes err, and it is prudent never to place too much trust in whatever has even once deceived us; and next because every night in our dreams we seem to observe or to imagine innumerable things which are non-existent {elsewhere}; and to a man thus doubting, there appear no indications by which he can distinguish sleep from waking with certainty {and can know whether the thoughts which come in dreams are more false than the others}.
The key bit is that “in our dreams we seem to observe or to imagine innumerable things,” a statement which is in fact suggestive of an imagination model. Descartes treats dreaming as being like perception (we seem to observe things), but doesn’t treat dreaming as perception in a way that would distinguish it from imagining (we imagine those things). Note, further, that it doesn’t matter for Descartes’s argument whether the dreamer forms false beliefs or just propositionally imagines: What matters is the lack of “indications” for the subject that would distinguish between dream experience and waking experience; that is, what matters for Descartes is that the experiences be phenomenologically similar such that at any given time you could not be sure on the basis of your experience that you were awake instead of dreaming. If this is correct, then the mature Descartes found in the Principles does not assume the orthodox theory of dreaming as Ichikawa states it and the dream skepticism argument that he gives is perfectly compatible with the imagination model of dreaming.
