Dennett, Daniel C. (1987) Commentary on Cam. Philosophy and Phenomenological Research XLVIII 339-341.

Commentary on Cam

Philosophy and Phenomenological Research, XLVIII, 339-341, December 1987.
Daniel C. Dennett
Center for Cognitive Studies
Tufts University

In "Propositions about Images" Philip Cam accurately analyzes and criticizes the grounds I gave, in the works he cites, for my denial that we have privileged access (of any sort) to anything deserving to be called a mental image. He shows that I did not deal properly with the question of how I would interpret the ostensive force of "this" and "that" in an introspective judgment of the sort: "Now it looks like this and now it looks like that." What can one be ostending or referring to in such a case, if not to an image (or some feature of an image)?

My reply in 1979, which he quotes, is that these demonstratives are short-hand "promissory notes" for hard-to- articulate descriptive propositions, and this is not very convincing. It isn't clear to Cam, he says, that my promissory notes can be repaid --without recourse at some point to a reference to images.

I think Cam is right, but now I see that I was wrong to put any importance on the possibility-in-principle of getting all the "content" in an introspective judgment "expressed" (or "explicit") in a sentence. That line of defense was doubly confused on my part. First, as I have more recently argued ("Beyond Belief," 1982), there is no uniform and problem-free understanding of the relation between propositions (the abstract entities taken traditionally as the "objects" of "propositional attitudes) and sentences (of English or any other natural language). You can't put into English the most "propositional" (least "imagistic") of "propositional attitudes" without risk of loss or distortion, so it should not surprise or dismay me that there is a difficulty in finding an intuitively adequate sentence to express an introspective judgment about the nature of one's putative mental images.

In the second place, I was wrong to think I needed such an escape to sentences to preserve my view about the unprivileged (or even underprivileged) access we might have to anything worth calling an image. All I needed was a distinction between images and judgments about images. (See "Two Approaches to Mental Images" chapter 10 in Brainstorms, 1978, where I first formulated this alternative idea.) In another more recent paper ("How to Study Human Consciousness Empirically: or, Nothing Comes to Mind," 1982), I developed the proposal that one's capacity to talk about what goes on in one's phenomenology might be unavoidably metaphorical, as well as unprivileged, but that this did not have the implication that there was any residual ineffable content that posed a challenge to materialism, or functionalism.

In those two 1982 papers I did not, however, explicitly draw attention to my abandonment of the position Cam criticizes here, and it is hardly obvious how my later view is supposed to replace my earlier view. The fundamental point I want to retain is that if there is, on some occasion, something going on in you that is well decribed as an image, your access to that is in no way more intimate or more direct than your access to "public" images--of the sort you might draw on paper.

Suppose you draw a picture on a piece of paper and then look at it. (It might be a duck-rabbit, for instance.) Now unless you are extraordinarily absentminded or drunk or otherwise addled, you are no doubt going to be in a very good position, epistemically, to describe the image you have created--but in no better position in principle than any third party who watched you create it. You may know things about what you were trying to do that the third party doesn't know, and you may know things about your subsequent authorial intentions undreamt of by the third party (e.g., you had intended to draw a duck-rabbit, but now that you look at it, it looks rather like an old man with a big nose looking up, so that is what you, the "artist", decide it is.) You are in a position, as author/artist, to declare an interpretation of the image you see, but that is the extent of your privilege.

Suppose now instead that you solve a problem "in your head" with the aid (so you think) of a mental image you create. Unless you are seriously--perhaps even systematically--misled about the nature of your activity (a prospect that must not be lightly dismissed, given the evidence of confabulation and illusion in a host of human cognitive activities), you are no doubt going to be in a good position, epistemically, to describe what you have done--though probably in somewhat metaphorical terms--and if what seems best to you to say is that you have created an image, that is apt to be a fruitful way for third parties to understand the phenomenon, but your "access" to your creation in this case is just like your access in the previous case.

It is tempting to suppose in the former case that there is another image, a private, subjective image in between the public image on the page and your various interpretive judgments about that public image. And there may well be; there may be something in your head that is rather like what is in your head in the latter case. But your judgments about that inner image are no more privileged than judgments about the public image--and your authorial privilege to stipulate an interpretation of the public image is no less secure than your privilege to stipulate an interpretation of your private images. Interpretations aren't images, however.

What does this leave me to say about someone who avers, in the relevant circumstances, "Now it looks like this and now it looks like that"? To what is this person referring? Perhaps, as Cam suggests, to some actually different images--though whether this is the best way of talking is not something for the subject to decide! If not to anything worth calling an image, then to some other "data structure" which permits a distinction to be drawn between duckish duck-rabbits and rabbitish duck-rabbits, and whether it is the sort of thing that lends itself to capturing in a sentence is an open (and not particularly important) question.


Dennett, D. C., 1982 "Beyond Belief" in A. Woodfield, ed., Thought and Object, Oxford Univ. Press, pp. 1-95.

Dennett, D. C., 1982. "How to Study Human Consciousness Empirically: or Nothing Comes to Mind" Synthese, 53, pp.159-80.