ARISTOTLE (384-322 b.c.e.)

Aristotle was born in northern Greece, in the town of Stagira, in 384 B.C.E. At seventeen, he went to Athens and became a student in Plato's Academy, where he remained for twenty years. Although greatly influenced by Plato and by the pre-Socratic philosophers, especially Empedocles, Aristotle was a highly original thinker and a disciple of no one. In 347 B.C.E. he left Athens and traveled extensively in Asia Minor, becoming tutor to Alexander the Great in 342 B.C.E. Seven years later he returned to Athens and began his own school, the Lyceum. After the death of Alexander the Great in 323 B.C.E., he left Athens, and he died the following year in Khalkís, a few miles north of Athens.

In his main work on memory, De memoria et reminiscentia, Aristotle tries to dissect out the central phenomena to be explained, and suggests mechanical explanations of a general sort to account for them. In his scientific works, Aristotle typically seeks the reality behind the appearances, and he expects that the reality may be different from what it seems. This is especially forward-looking in the case of mental phenomena, where subsequent thinkers, such as René Descartes (in the seventeenth century) and Zeno Vendler (in the late twentieth century), insist that mental reality must be exactly as it seems. Aristotle's collection of memory phenomena displays some systematicity, and with characteristic insight, he lights on several basically correct classifications. Nevertheless, to modern eyes some of his collection is a bit of a jumble, and the mechanical explanations tendered are so implausible that they must have been no more than helpful metaphors to him.

Aristotle's relentlessly naturalistic perspective, however, gives him a decidedly modern stamp. That is, he sought physical rather than supernatural or spiritual explanations for memory phenomena, and he well knew the importance of observations even though his own were occasionally mere assumptions. (For example, he thought women had fewer teeth than men.) In the absence of a developed biology, experimental psychology, or neuroscience, he could hardly be expected either to envisage explanations in terms of neuronal connectivity or to know how to penetrate learning phenomena at the behavioral level.

Observations and Explanations

In commenting upon memory and learning phenomena, Aristotle's fundamental distinction is between recalling information to mind and storing information, or, as he puts it, between remembering, which is "the reinstatement in consciousness of something that was there before" (451b6), and memory, "the existence, potentially, in the mind" (452a10), of an earlier perception or conception. In modern parlance, this is the distinction between remembering in the occurrent sense and remembering in the "stored" or dispositional sense. The central problems, in Aristotle's view, are to explain three things: 1. how a perception of a state of affairs can be stored, 2. how it can be brought to mind later, and 3. how it happens that, when it is brought to mind, the relation between the representation and the original state of affairs, now absent, is such that the first is a memory of the second and is known to be such. In contemporary dress, these are the problems of information storage, information retrieval, and the general problem of how representations represent.

Aristotle tries to explain information storage by appeal to the analogy of imprinting soft wax with a seal. He reasons that sense perception is somehow like a picture and that it is the perception picture that stamps its likeness to create a memory. Apparently the perception is stamped on the soul (Aristotle has a physical, not a supernatural, conception of the soul), or at any rate, it is stamped on some sort of physical stuff that can be in causal interaction with it and can take on some of its properties. This helps address the representation problem. The imprint (memory representation) resembles, physically, the perception (perceptual representation), which in turn resembles, physically, that of which it is a perception. So by transitivity of resemblance, there is a correlation between stored representation and original state of affairs. Aristotle's conclusion that there must be a resemblance was taken as axiomatic by most subsequent thinkers, and they searched for the parameters of physical resemblance. Research since the 1970s, especially in computer science and neuroscience, has revealed that representation does not require resemblance in any straightforward sense, a radical departure from earlier theories.

In asking how representations represent, Aristotle identified a truly fundamental problem. Still only partially solved, it remains a central problem, though it is now addressed within the framework of modern psychology, philosophy, neuroscience, and computer science.

Understanding the importance of broad systematicity in a theory, Aristotle tests a theory's strength by seeing how much can be encompassed within its ambit. Thus he claims that the stuff that receives the imprint may have varying degrees of imprintability. Explanations are then forthcoming for one's poor recollection of early childhood and for declining memory in the elderly: In very young children the stuff is too much like running water to take the imprint; in older humans, the stuff hardens and no longer is very impressible. Extending this idea further, Aristotle thinks a related explanation will apply to his observation that those who are "too quick" and those who are "too slow" also have poor memories. Exactly what phenomenon he is addressing here is unclear, and this may be one of those inexplicable Aristotelian "observations" that need a much broadened base of data.

The representation problem, Aristotle notices, has a further dimension. When an image from memory comes to mind, how do we know that it is a memory, rather than a thought or image without relation to bygone events? That is, how does the occurrent presentation carry the information that it is a memory? His answer has two parts. First, sometimes we do get confused, and we think a presentation is a memory when it is not (false memory); and sometimes we have a memory presentation but are unaware that it is a memory. So the system is imperfect. Second, when the system does work, it is because for animals with memory, "the organ whereby they perceive time is also that whereby they remember" (449b30). The idea here is that when perceptions are stored as memories, they are also somehow indexed as to time, so that the imprint bears not only the perception's shape but also its "whenness."

Retrieval appears to require something like an image or an iconic presentation that resembles the original perception. The mechanism of retrieval should, one surmises, have to do with something taking up the stored imprint and re-presenting it, but in fact Aristotle says nothing of this. Instead he discusses the phenomenon of association, noting that events experienced together are often remembered together. He explains associated recollections by saying that the "movement" of a perception causes the "movement" of the memory. He sees, therefore, that part of the theory of storage will include the relations between associated memories, but he neither provides an account of those storage relations nor elaborates on how information is retrieved by the "movements" (451b15-30).

In Historia animalium Aristotle suggests that humans and animals differ in that humans alone can remember something at will (488b25), though he also notes in De memoria et reminiscentia that recollections can occur without effort. Indeed, he observes that melancholics often have obsessive memories, try though they might to repress them. In the physicalist spirit, he conjectures that melancholics have more moisture around their sense perception center, which is easily set in motion, thus explaining the memory's being presented again and again despite one's will.

Aristotle believed that animals differ in whether they have the capacity to store their perceptions; animals with the capacity to do so have genuine knowledge of their world, whereas animals lacking the capacity merely respond to their current perceptions on the basis of their innate dispositions. The advantage of storing perceptions is that the stored items may come to have systematic relations among themselves, with the result that the animal can recognize different individuals as belonging to the same category. In humans this means, for example, that a pine tree, a yew, and an olive tree may all be recognized as similar despite differences in shape, size, and color. He says that the soul is so constituted that the universal "tree" can be developed from the stored perceptions of individually distinct items. A slug, on the other hand, lacks the capacity to generalize across individuals because it lacks the capacity to store information.

In Aristotle's view, storing information provides the similarity substructure that underpins both scientific categorization and the skilled knowledge displayed by craftsmen who can make many different clay pots or ship's captains who can sail under many different conditions. In modern guise, his idea is that generalization to items that are relevantly similar but incidentally different, both perceptually and behaviorally, requires information storage. Additionally, he regards this capacity as enabling experience, the reason being that experience requires understanding, which in turn requires categorization of perceptions. Consequently, animals such as humans have genuine experience; animals such as slugs do not (Posterior Analytics, 99b36;l00a5).

Conclusion

Any inclination to feel smug about Aristotle's shortcomings should be tempered by noting that even current classifications of learning phenomena are controversial and tentative, and experimental psychologists are sometimes chided for doing little more than codifying common sense. Nor, of course, should Aristotle himself be blamed for the slavish adoption of his every word by uncritical monks in the Middle Ages. Aristotle the scientist-philosopher was anything but dogmatic. Twentieth-century physical explanations—although not mechanical, but electrical and biochemical—sit well with his abiding naturalism.

For a long period in the history of thought, Aristotle's views on nearly everything were taken as authoritative. His Metaphysics probably had the greatest impact; however, the work on memory was not especially influential.

Bibliography

Aristotle (1941). De anima, trans. J. A. Smith. In Richard Mc Keon, ed., The basic works of Aristotle. New York: Random House.

—— (1941). De memoria et reminiscentia, trans. by J. I. Beare. In Richard McKeon, ed., The basic works of Aristotle. New York: Random House.

—— (1941). Historia animalium. In Richard McKeon, ed., The basic works of Aristotle. New York: Random House.

—— (1975). Aristotle's Posterior analytics, trans. J. Barnes. Oxford: Clarendon Press.

Beare, J. I. (1906). Greek theories of elementary cognition. Oxford: Clarendon Press.

Churchland, P. S. (2002). Brain-wise: Studies in neurophilosophy. Cambridge, MA: MIT Press.

Descartes, R. (1649; reprint 1968). Les passions de l'âme. English translation in E. S. Haldane and G. R. T. Ross, trans. (1911), The philosophical works of Descartes. Cambridge, UK: Cambridge University Press.

Edel, A. (1982). Aristotle and his philosophy. Chapel Hill: University of North Carolina Press.

Kahn, C. H. (1966). Sensation and consciousness in Aristotle's psychology. Archiv für Geschichte der Philosophie 48, 43-81.

Ross, W. D. (1923; reprint 1959). Aristotle. Cleveland: Meridian Books.

—— (1955). Aristotle's "Parva naturalia." Oxford: Oxford University Press.

Sorabji, R. (1972). Aristotle on memory. London: Trinity Press.

Patricia SmithChurchland

GeorgiosAnagnostopoulos

Revised byPatricia SmithChurchland