2. The intentionality of memory

This week, I’m writing a series of posts on my new book Memory: A Self-Referential Account (Oxford University Press, 2019). The post today concerns the content of memories; the topic of chapter three.

I defend the view that memories are self-referential in that they represent a property of themselves. Specifically, they represent themselves as having a particular causal origin. I motivate this view by highlighting the difficulties for alternative views. I assume a notion of content according to which the content of a memory is constituted by its truth conditions. And I use a series of thought experiments in which I test our intuitions about the veridicality of a memory with regards to some possible situations in order to attribute several elements to the content of that memory.

It goes like this. Suppose that I look at a red apple and I have a perceptual experience that presents it to me as being red. Let us call it ‘P’. And suppose that, a few days later, I have a memory that originates in P; a memory that I would express by saying that I remember a red apple in front of me. Let us call it ‘M’. What is the content of M? The most natural view is that, given the way in which I express my memory, the content of M is that there was a red apple in front of me. But this would be true of a situation in which I never experienced the apple perceptually. And yet, it intuitively seems that, by having M, I am not accurately representing such a situation. This suggests that our memories provide us with information about our own past experience of the world.

In light of this outcome, one might be tempted by the view that the content of M is that I perceptually experienced a red apple in front of me. But this would be true of a situation in which, let us say, I misperceived a green apple as being red. And yet, it intuitively seems that, by having M, I am not accurately representing such a situation. This suggests that our memories provide us with information about the objective state of the world in the past.

Let us suppose, then, that the content of M is that I perceived (veridically, that is) a red apple by having P. This would not be true of either of the two situations entertained above. But it would be true of a situation in which, after perceiving the red apple, I completely forget about it. And yet, intuitively, this is not a situation that I accurately represent when I have M. This suggests that our memories provide us with information not only about the objective state of the world and our perceptual experiences in the past, but also about the memories themselves. They provide us with the information that they come from particular perceptions in the past.

Thus, I conclude that the content of M is that M is due to the fact that I perceived a red apple by having P. The thought is that memories wear, so to speak, their causal histories on their sleeves. This seems to take care of our intuitions about the veridicality of M with regards to the three possible situations sketched above.

Now, let us zoom out a little, and see how this view about the content of memories fits with the view about the metaphysics of memory discussed in the first post. According to that view, episodes of remembering were to be construed functionally. Thus, remembering a red apple consisted in having some mental image or another which tends to be caused in me by a perceptual experience of a red apple, and which tends to cause in me both the belief that there was a red apple in front of me, and the belief that I experienced this to be the case in the past. When I remember a red apple, I have a mental image with that functional role. And that mental image is what we call a memory. The view about the intentionality of memory is, now, a view about the content that those mental images have.

But I am not suggesting that all and only memories have self-referential contents of the kind I’m describing above. Maybe episodes of imagination can have those contents. (Maybe this is why we mistake them for memories sometimes.) And maybe some memories can lack those contents. (We will see an example of this type of case in an upcoming post.) After all, the contents of memories is not what makes them memories; their functional role is.