Aristotle on the Nature of Memory

How does Aristotle conceive of memory? Why does Aristotle’s theory of memory have such broad implications for his philosophy of mind, and for the philosophy of mind as a whole? This article begins with a discussion of the defining features of Aristotle’s theory of memory — in particular, the relationship between Aristotle’s theory of memory and the mental image. The way in which Aristotle distinguishes memory from recollection is then discussed, and the article then concludes with an evaluation of the implications of this distinction in terms of how successful Aristotle’s attempt to naturalize elements of the mind — that is, to explain them in terms of corresponding parts of the body and their processes — turns out to be.

Aristotle begins his discussion of memory in On Memory and Reminiscence by attempting to distinguish memory from other, apparently similar or related faculties.

First, Aristotle holds that remembering must be strictly distinguished from the original processes by which a certain sensation that we are recalling was first felt. In other words, memory is not a total recreation of the conditions of experience, and must therefore be seen as a process with its own causal context to be explored.

Aristotle also claims that memory, in contrast to sensation, involves an awareness of time. It is an intrinsic quality of memory that it is past, that the temporal conditions, as well as the spatial content of the experience of memory, are fixed parts of it. Yet, of course, memory is clearly bound up in our faculties of perception, and an attempt to clearly distinguish memory from experience necessarily raises a host of further questions concerning these relationships.

Some of what Aristotle has to say about memory appears to abnegate responsibility for characterizing these relationships: “Memory, then, is neither sensation nor conception, but a state of having one of these or an affection resulting from one of these, when some time elapses.” Is memory a kind of simulated experience, a kind of experience without a cause? This can’t be true, because Aristotle makes it clear that memory — like perception — has an organ the alteration of which is responsible for it.

Perception and Memory 

So what is the relationship between our perception of certain sensations or our having certain conceptions, and our memories thereof? The concept Aristotle turns to is that of the imagination, though the relationship between the imagination and memory is disputed.

It seems that Aristotle turns to speak about how the imagination relates to the various other senses on the basis that the imagination and memory are, in whatever way, closely related. On Memory proceeds from the assumption that thought requires images, that images are necessary conditions of thought itself, and that the imagination is the faculty by which images produced by our previous perceptions of sensation are preserved, stored, and ready for deployment in thought.

Aristotle here seems ambiguous between two possible understandings of the way in which imagination, thought, and perception relate. On the one hand, he might be saying that the imagination is that which facilitates the rearrangement of previous perceptions (those original perceptions of sensation) into the new thing that constitutes thought. On the other, he might be saying that the imagination reiterates images exactly as we first perceived them.

In this latter interpretation, we are presumably to think of the imagination as almost entirely conflated with memory. In this case, the only difference between imagination and memory is the sense of temporal location, which is intrinsic to memory but not to the imagination.

Aristotle himself summarizes the relationship between memory and imagination as follows:

“and (1) those things that are essentially memorable are also those of which there is imagination, (2) while those that are accidentally memorable are those that do not occur without imagination.”

This is an even more tantalizing picture of the relationship — what is it for something to occur without imagination? Can something rely on the imagination without itself being of it?

If the essential concept of the imagination and of memory is the image, we should then ask what the image really is for Aristotle. To what extent can a mental image be given a physical explanation, as Aristotle is inclined to give of perception itself?

Even if there is a physical component to the image insofar as it is related to Aristotle’s theory of perception, they are on some level non-physical as well. Thoughts occur through the body, but they reside in the soul (which, for Aristotle, subsumes all of the functions which we would today call “mental”). Yet there remains some degree of scholarly controversy about how we should conceptualize the relationship between the body and soul which allows this kind of partial physicality.

The question of whether images are pictorial or non-pictorial has similarly prompted serious debate, as has the very possibility of there being a non-pictorial image in the first place. For what it’s worth, the evidence in favor of Aristotle holding a non-pictorial view of the mental image seems far less decisive than the one in favor of him adopting the opposite view.

Images and the Mind 

If memory is an image which arises in the moment of remembering (in other words, in the present), then in what sense is memory intrinsically marked by the past? The thing with which the memory is concerned, the thing which it focuses on, is in the present, even if the thing which corresponds to it (the object or thing of which the image is an image of) is not. In other words, the question of how our experience of memory is intrinsically linked to the past depends very much on the relationship between an image and its corresponding object.

Aristotle draws a distinction between images that are images of something else, and those which are not, and draws a further distinction between images that are of something else as it was in the past and those which are not. This is curious, however, given that the condition for what counted as a memory was given earlier as to do with its intrinsic relation to the past. Now it sounds like memory is being defined according to a certain way of categorizing images.

Moving on to the distinction implied in the title of Aristotle’s work, Aristotle posits that there is a subtle but significant difference between memory and recollection. The distinction relates the faculty of memory not only to that of perception, but to that of understanding, mind, or, to use the Greek term which is still in use among philosophers, nous.

What is this difference? Part of the difference is to do with immediacy: memory is not a kind of instantaneous response to experience, whereas recollection is instantaneous. The reason Aristotle gives for this differentiation on the grounds of immediacy is that recollection is an active process. Even if it is not a strictly intentional one (it isn’t hard to imagine involuntary recollections), it very often does imply a kind of interior search, a kind of decision being taken with respect to which parts of what we can recall are needed at a given moment.

This conceptual distinction is so critically important within Aristotle’s framework because of the corresponding distinction in nature, or more specifically, in Aristotle’s conception of the human body. Whereas memory does have a corresponding organ, recollection does not, because recollection is an element of the intellect, mind, or nous. It is a part of our faculty of understanding.

Here Aristotle’s theory of memory has broader consequences, both for Aristotle’s theory of mind as a whole and for the philosophy of mind in general. This is an interesting element of Aristotle’s theory of mind more generally, because it somewhat distorts, or at least qualifies, the ordinary conception of his theory as one in which the structure of the concepts of mind is literally reflected in the structure of the body (that is, the corresponding organs).

Here, two elements of our mental life that are self-evidently closely related are attributed to totally different organs. We might think that this seems emblematic of the tendency of our conception of mind to resist whatever attempt we make to corral and confine it to a straightforward correspondence with specific physical locations and processes.

If one still wants to claim that there will be corresponding physical entities and processes for every mental process, it could be said that we have such an imperfect understanding of the mind and of those elements of our body which might constitute it that this is not a definitive response.

It is tempting to reply to this counter-argument with a simple dilemma, with which this article will end: is there a fixed point at which we will understand the mind and body well enough to determine whether every mental process has a corresponding physical component? Or is the theory of physical correspondence effectively irrefutable because we will never know when we have come to a point of understanding the body and mind well enough to assess it?