Plato’s four premises (A, B, D, and F) are insecure for various reasons. Premise A: The appearances of things do change. 1. Is every change equally significant here? Is there a difference between adjustments, growth, cycles, and flip-flops? 2. Couldn’t a change be part of a permanent (or semi-permanent) pattern? Premise B: The essences of things don’t change. 1. Do things (or even kinds of things) actually have “essences”? 2. Why should what changes be necessarily inferior to what doesn’t? Unless something is perfect, why couldn’t it change for the better? 3. Why couldn’t change be part of the essence of some things? Premise D: The reality of things resides in their essences, not in their appearances. 1. If this (as well as premise B) is so, then only unchangeable things are “real.” 2. But if we have any reason to think that something is real (say an ocean wave) that does change (as ocean waves do), then we have reason to think this premise (or premise B, or both) is not so. Premise F: Appearances are all that we have sensory access to. 1. This may be so, but (in fact) many things to which we have sensory access are remarkably constant, such as the effects (appearances) of gravity, for example. 2. Thus, if this premise is so, we have some reason to dispute premise A.
Plato’s metaphysical proposal (J) and its dependent hypotheses (K and L) constitute a nice exercise in hypothesis construction and modeling. But this particular construction presupposes that there are Forms, that the mind is substantially different from the body, that the mind does preexist the body, that in that preexistence the mind has access to the Forms, and that what it intuits directly there, it can remember here. A. What reasons are there, if any, for thinking that any of these presuppositions is true? B. What grounds are there for determining the merits of such a hypothesis and model? 1. Can we directly discover what the world itself is like? 2. If so, can we lay this model up against the world and see whether it is accurate? 3. If we can’t do that, are there pragmatic advantages or disadvantages in using this model rather than some other model?
Although reason does require content to process, it is not clear that suppositional recollections from a past life (that we may not have had and, a fortiori, may not have any recollections of) are the kind of content it needs. To fill in the gaps in that needed content, sensory experience is positively attractive, for all its alleged fallibility.
In any case, intuition, memory, and reason are often unreliable themselves. A. Every intuition has a rival. There are even rival geometrical axioms. B. I misremember things all the time, and I don’t think I am unique in this. C. One has to learn to reason accurately. It is uncommon and not innate. D. If all these things and sense experience are fatally unreliable, then the door is open to systematic skepticism of the sort Peter Unger advocates (unless we can find some external guarantor, as Descartes claimed to find in God).
Furthermore, like all dualisms, this scheme generates a methexis problem. A. Methexis means “interaction” or “participation.” Descartes has a methexis problem about the interaction of the mind and the body. The ancient Greeks had a methexis problem about the interaction of the good and humans. Traditional Christians have a methexis problem about the incarnation. B. The methexis problem here is that if we (now) live in the world of appearances, not in the world of the Forms, how can memory “bridge the gap”? Where does memory live?
Finally, this view encourages us to extrapolate mathematical thinking beyond its bounds. A. Plato’s view suggests a rather “geometrical” model for belief and knowledge. (In geometry, self-evident assumptions are said to yield deductively certain conclusions.) B. In so doing, it does not take into account the possible sterility of purely formal truths and the availability of alternative “axioms.” C. It ignores the difference between what is formally true within a system and what is descriptively true of the world. D. Even Euclidean geometry is contingent. The work of Lobachevsky and other mathematicians in the 20th century showed that alternative geometries are possible. E. Although this model may work nicely for figuring out the properties of Euclidean triangles, it is problematic if we try to extrapolate it to wider realms.
In spite of the fact that Plato’s view makes good use of three basic tools of thinking—intuition to provide input, memory to provide continuity, and reason (to uncover and manipulate what we intuit and remember)—it is inadequate. Perhaps readmitting sense experience will help.
[Courtesy: Professor James Hall]