Strong dialectic of the sort practised in first philosophy must justify the common beliefs, or most of them, by appeal to some independently reasonable principles. In De Anima i Aristotle presents the common beliefs, and the puzzles they raise. As usual, it is worth examining the puzzles, to see that the questions have not been stated correctly. We agree at the outset that the soul is ‘some sort of principle (archê) of animals’ (402a6–7); but this extent of agreement does not take us far.2
Do all souls belong to the same species or not? If not, do they differ 〈only〉 in species, or 〈also〉 in genus? As things are, those who discuss and investigate the soul would seem to 〈ignore this question since they〉 examine only the human soul. Nor should we forget to ask whether there is just one account of soul, as there is of animal, or a different account for each type of soul—e.g. of horse, dog, human being, god. 〈On this second view〉, the universal animal is either nothing or else it is posterior to these; and the same will apply to any other common thing predicated. (402b1–9)3
Even if the soul is some sort of principle, it may not be worth investigating in its own right; for things with souls may not (as such) have any genuine or important common property.
To find out whether having a soul is an important common property we turn to its commonly accepted attributes. ‘All define the soul by three things, one might say—by movement, by perception, and by being incorporeal’ (405b11–12).4 Such different beliefs about the essential features of soul might well suggest that different theorists are really talking about different things. We cannot simply combine different answers and say that the soul is the cause of both perception and movement; for if the cause of perception and the cause of movement have nothing more in common, our use of the term ‘soul’ for them will not show that having a soul is an important common property.5 It will be especially hard to see any common property if we follow the conception of cause that previous theorists have assumed. For they seek material causes, and different material causes seem appropriate for understanding movement and perception.6
Aristotle even suggests, contrary to common beliefs, that the soul is the principle of life, not simply of animal life, and hence that plants as well as animals have souls.7 If we agree that everything that is alive has a soul, we cannot even assume that the soul is essentially the cause of movement or perception. If its functions must include the nutritive and reproductive functions of plants, it seems even clearer that having a soul is not a genuine common property, but a mere disjunction of properties.
So far we might assume general agreement that the soul is some sort of material stuff. But Aristotle denies even this degreeofagreement;forheclaimsthatcommonbeliefsalsoassociatethesoulwiththeincorporeal.8 Wethinkthatin some way an organism is one because it has one soul (410b10–15); and no material stuff seems to explain this role. For if we identify the soul with the stuff that causes movement or perception, we want to know what makes that stuff and the rest of the material components into one organism; and the stuff that causes movement and perception does not answer this question (411b5–14). If the soul were simply a material stuff distributed through parts of the body, then we would expect to be able to split the body into different parts, and to find a part of the soul in each; but we do not think thisispossible(411b14–30)9 Webelievethesoulisindivisiblypresentinthewholeorganism;andthisbeliefishardto explain if we identify the soul with some type of material constituent.
We connect the soul not only with the unity of organisms at a time, but also with their identity and persistence through time. We think Socrates exists as long as his soul exists, that without the soul he perishes (411b7–9), and that if one soul replaces another one person has perished and been replaced by another. But we find it hard to see how this feature of the soul could be explained by some particular material stuff of the sort that might seem to explain perception and movement.10
If a corporeal soul seems not to explain some features of the soul, we might conceive an incorporeal soul in different ways. First, we might argue that if it cannot be a material substance, it must be an immaterial substance, neither identical to nor constituted by a material body. This is the Platonic solution. Aristotle argues that it underestimates the close connexions between the soul and the body whose potentialities it actualizes (407b13–26)11
Alternatively, if some functions of the soul seem not to be functions of a material stuff, but we think the soul can be a substance only if it is a sort of material stuff, we infer that the soul is not a substance at all, but some arrangement or ‘attunement’ (harmonia) of material constituents. This theory shows that we can recognize the non-material character of the soul without treating it as an immaterial substance. None the less, Aristotle rejects any such theory, (407b27–408a28), maintaining that the soul is a substance.12
The survey of common beliefs about soul shows that a general account of the soul cannot be read off them, and that we cannot even expect to solve the puzzles they raise by a direct attack. For it is not clear that the apparent disagreements (about whether the soul is material or not, and about what kind of material it is) are really disagreements about the answer to one question. If questions about the soul are really just questions about the cause of perception and motion, and about the source of unity in the subject, then different answers need not be contradictory. In this case, as in Metaphysics iii, Aristotle's predecessors have overlooked some important preliminary questions; for we can hardly settle their apparent or real disagreements without deciding what we should expect an account of the soul to be like.