Tuesday, May 18, 2010

Successful Theistic Explanations (Part 2) - Background Knowledge and Past Success


This post is part of my series on Chapter 7 of Gregory Dawes's book Theism and Explanation. In this chapter, Dawes's assesses the merits of theistic explanations in light of a series of six explanatory virtues.

In part 1, I covered the general background to chapter 7 and looked at the first of the explanatory virtues (testability). In this part, I will cover the next two explanatory virtues: consistency with background knowledge and past explanatory success.

To better appreciate this discussion it is worth remembering two things. First, a potential explanation is one that fits into an abductive schema. Second, a potential theistic explanation is one that appeals to divine intentions. In other words, it is an explanation that attributes events and states of affairs in the world to the goals of a divine agent.


1. What is Background Knowledge?
The first question we need to ask is: what is background knowledge? The answer is straightforward: background knowledge is any fact or theory, excluding the fact(s) we are currently trying to explain, of which we already have knowledge.

All things being equal, consistency with background knowledge is an explanatory virtue. As we shall see in a moment, some argue that in the case of theism things are not equal. For now, let's assume that they are.




One good illustration of the virtue of consistency with background knowledge is Darwin's theory of natural selection. When Darwin proposed his theory we had no way of observing the process of natural selection in action. So Darwin illustrated how his theory was consistent with another well-known process, namely: artificial selection.

The fact that Darwin's theory was consistent with artificial selection counted in its favour.


2. Is Theism Consistent with Background Knowledge?
Now we need to ask whether theistic explanations are consistent with background knowledge. On the one hand, given that we employ intentional explanations all the time, there is reason to think they are consistent. On the other hand, the nature of the divine agent is so completely different to the intentional agents with which we normally deal that the consistency is more apparent than real.

The main problem is that God is a non-physical, eternal, omniscient agent. The idea that such an agent could will into existence the physical, non-eternal universe is completely alien from our background knowledge.

Note that this is true even if you embrace some form of dualism about the human mind. J.L. Mackie made this point well when he criticised Swinburne's cosmological argument. He said:
All our knowledge of intention-fulfillment is of embodied intentions being fulfilled indirectly by way of bodily changes and movements which are causally related to the intended result, and where the ability thus to fulfil intentions itself has a causal history, either of evolutionary development or of learning or of both. (From The Miracle of Theism, p. 100)
Mackie's statements are consistent with the idea of a non-physical human mind, and they signal how alien the divine mind really is.


3. Is Background Knowledge Relevant?
Given the inconsistency of theism with background knowledge, theists might be inclined to argue that this particular explanatory virtue is not relevant when considering the explanatory merits of theism. Indeed, this is precisely what Richard Swinburne does and his argument is worthy of our attention.

Swinburne thinks that consistency with background knowledge is usually a relevant consideration but that it becomes less and less relevant as the scope (or breadth) of the explanation increases. The reason for this is that as an explanation explains more and more facts, there is less and less independent background knowledge. And since some theistic explanations are extremely broad in scope, it follows that there is practically no background knowledge with which they can be consistent. The only exception is logical knowledge (and some theists think god explains that as well!).

At first glance this seems like a sound objection, but it falls apart on closer inspection. What Swinburne is saying here is that theistic explanations are so broad that they can explain all true propositions (P1, P2....Pn) about the world. But the reality is that no theist does, or even could, offer such a broad explanation.

This is best illustrated by taking a look at cosmological arguments. These are the potential theistic explanations with the broadest scope. But even then they only cover general propositions about the world and not the set of all true propositions (P1...Pn). They cover propositions like "there cannot be an infinite set of contingent entities", "there cannot be an infinite sequence of causes", "there cannot be an infinite sequence of temporal events" and so on. Each of these propositions is only part of the set of all true propositions.

So, it is always going to be possible to distinguish between what is being explained by theism, and background knowledge that is being taken for granted. And given that it is possible to do this, theism should be consistent with background knowledge (if it wants to be taken seriously as an explanatory hypothesis).


4. Past Explanatory Success
This brings us to the question of past explanatory success. In some ways this is similar to the question of consistency with background knowledge, but Dawes thinks it is significant enough to be considered a separate explanatory virtue.

Past explanatory success refers to the track record of a particular explanatory hypothesis. For example, the reason why so many people have found evolutionary theories of human psychology to be appealing is that evolutionary explanations have a good track record outside of human psychology. There may, of course, be other reasons to discount evolutionary psychology (lack or failure of testability for example).

The point can be generalised: the past explanatory success of naturalistic scientific explanations is a reason to favour them over supernatural explanations (such as theism). This is not simply an illegitimate attempt to stack the deck against theistic explanations. The successes of naturalism are numerous and real. We are aware of them every time we take an antibiotic or turn on a laptop.

Until theistic or supernatural explanations have had similarly productive explanatory successes, there will always be reason to be sceptical of them.


That's it for now. In the next part of this series we will look at the explanatory virtues of simplicity and ontological economy.

No comments:

Post a Comment