Thursday, February 24, 2005

Another Look at a Reppertian Argument

Victor Reppert gives the following argument in Philosophia Christi, vol. 3, no. 1 (2003), p. 16:

1. If naturalism is true, then no states of the person can be either true or false.
2. Some states of the person can be either true or false.
Therefore,
3. Naturalism is false.

Before reexamining this argument, let me say that Victor and I are 'on the same team': we are both resolute anti-naturalists. Thus our disagreement, if disagreement it is, is intramural: within the walls of the anti-naturalistic encampment.

My criticism is simply that a sophisticated naturalist will not feel compelled to accept the premises. But first a very minor logical quibble. Strictly speaking, (3) does not follow from (1) and (2). What follows is: Naturalism is not true. The assumption of Bivalence -- which of course I grant -- is necessary to arrive validly at (3) by Modus Tollens.

That quibble aside, why should a sophisticated naturalist feel compelled to accept (1)?

A necessary condition of being a naturalist is the insistence that mind is ontologically dependent (dependent in its existence) upon the natural world. The dependence can be spelled out in a variety of ways giving rise to a variety of naturalisms. Some naturalists say that mind is epiphenomenal, others that it is emergent, still others that it is supervenient, while the most extreme of all say that it is reductively identifiable with brain processes ( Identity Theory) or with overt behavior and behavioral dispositions (Behaviorism). The naturalists’ main point, however, is that mind cannot exist without a material substratum of a certain complexity and organization: there are no disembodied minds, hence no God as classically conceived, no souls capable of pre-existing or post-existing their embodiment, no angels or demons.

Suppose minds or persons are supervenient. Why, given naturalism, couldn't some states of persons be either true or false? Reppert's (1) may be true for a naturalist who is an identity theorist, but not for one who is a supervenientist.

But suppose our naturalist is an identity theorist. Premise (2) appears to beg the question against him. For if mental states are brain states, then they cannot be true or false. The identity theorist could still find a place for truth in his world by identifying truth-bearers not with states of mind/person, but with some species of abstract object like Fregean propositions.

My view is a that a viable naturalism must admit nonphysical truth-bearers. Here is an argument.

Surely there are truths that hold at times when there are no minds. The thesis of naturalism itself, if true, is such a truth. At the time of the Big Bang, and for a good long time thereafter, natural conditions were not such as to permit the existence of any minds; yet naturalism, if true, was true during this period. It was true during this period that nothing could be a mind unless it were based in matter of the right organization and complexity. Naturalism did not first become true when minds first emerged. As a truth about minds, it is true, if it is true, independently of any and all minds. And the same goes for infinities of other truths that do not depend on the existence of minds. Among these are truths about the physical universe and its parts.

Thus it was true before any minds evolved on earth that the earth was a spheroid with one moon, possessed an atmosphere of such-and-such a composition, etc. There are also necessary truths to contend with: they are true in every possible world, including worlds in which minds do not exist. Thus it is quite clear that the existence of truths in general cannot depend on the existence of the sorts of minds that naturalists will allow into their ontologies, namely, minds whose very existence depends on the existence of highly organized configurations of matter. (The special case of truths about minds is consistent with this general point.)

It follows that a sophisticated naturalist will not grant that some mental states are primary truth-bearers. The reason, again, is that there must be primary truth-bearers at times and in possible worlds in which minds do not exist. Such a naturalist will not grant the classical thesis of Thomas Aquinas et al. according to which truths reside only in minds to the extent that they correspond to external reality, as in the scholastic dictum, Veritas est adaequatio intellectus et rei. If truth consists in a correspondence between mind and reality, then mind must exist if truth is to exist. Theists such as Augustine and Aquinas will use this point to argue for the existence of a necessarily existent infinite mind. The sophisticated naturalist will hold instead that primary truth-bearers are some species of abstract object such as Fregean propositions that exist in sublime independence of any and all minds. This does not imply that the sophisticated naturalist cannot say of beliefs that they are true or false; it implies merely that he cannot attach these predicates to beliefs in the primary sense.

No sophisticated naturalist should lose any sleep over Reppert’s argument. A convincing argument from truth against naturalism must therefore show that the primary truth-bearers cannot be abstract objects. What I would argue is that primary truth-bearers cannot be abstract objects any more than they can be physical objects; they must be states of mind. Since this cannot be allowed by any naturalist who understands that some truths are independent of matter-based minds, naturalism is false. And since there are truths that outrun our states of mind, there must be an absolute mind.

But I can't ignore VR's question about how truth, if situated at the level of abstract objects, is relevant to the concrete occurrence of beliefs. This is a large topic, but why couldn't a naturalist who is a supervenientist say that some supervenient mental states inherit their truth-values from Fregean propositions? Of course, there is the problem of how a supervenient state can have any causal impact on physical states. Isn't Monokroussos working on that problem?