(In)coherent Mumbling

This post is part of an ongoing dialog between my friend Ernie and me about the validity of Christian belief.

At the end of my last post I left unanswered a set of questions that Ernie had posted. Tonight I intend to clarify some of what I said toward the end of Practical Truth as well as answer those questions.

First, I need to clarify this paragraph:

Ernie also asks "on what basis can we assert that science is really correlated with truth, as opposed to just a 'comfortable myth, held by a self-selected group, which seems to work much of the time'?" My answer is pragmatic: who cares? If a theory works (and that is the crucial qualification), it is useful and if another theory works better, it is more useful. I do think that science is correlated with truth in the realms it can address, simply because it works so well, but there is not a necessary causal relationship between "works" and "true", a point that came up yesterday in relation to the historical success of Christianity.

The question "who cares?" was a little too superficial, too flippant. Instead, the point that I failed to communicate is that we can never be sure that we have discovered the truth. We are fallible, both in observation and in rationality, and often we operate with incomplete information so that even if we observed and reasoned perfectly, we still could not be sure of everything. Rather than get all worked up over that limitation, we do the best we can, and for science the best we have done (so far) is to figure out what works, that is, what theories correlate with observations and which make verifiable (and verified) predictions.

That was science, though. What about other areas of inquiry? What about history? What about theology?

In history, it is still possible to make predictions, for instance about the nature of as yet undiscovered evidence. Beyond that, or rather, below that is the idea of coherency. We expect our historical assertions to be mutually compatible as well as compatible with that evidence we do possess. Coherency is more than logical compatibility though; to me it implies a kind of parsimony and abstraction. It "compresses" the relevant data into a more succinct form.

When there are conflicts among our assertions, we know that something is wrong, but not necessarily what. There may also be gaps, subjects for which no assertions are yet made. We can easily reduce the conflicts by allowing the gaps to grow; the challenge is to reduce the size of the gaps without increasing the conflicts, or better yet reducing the conflicts. This explanation is getting pretty close to what I described in Belief Optimization several months ago. One thing that I touched on at the end but must emphasize here is that so far as coherency goes, there may be multiple partially coherent (but not perfectly coherent) belief systems that assert different things. Which one is closest to the truth? How can we know? Coherency (or at least compatibility) is a necessary but not sufficient condition for being confident in the truth of our assertions.

For one example, consider the question of the resurrection. Did Jesus rise from the dead? Can we know that? Without going into any significant detail, let me contrast two different positions. First, to a theist the possibility that Jesus was raised seems not at all unreasonable, so given some reports that he was raised and the lack of any contemporaneous contradictory reports, a belief in the resurrection is not incoherent. On the other hand, to one with no prior belief in the supernatural, the possibility that the accounts of his resurrection were fictional is likewise compatible with the lack of any contemporaneous contradictory reports, so this (dis)belief of the resurrection is likewise not incoherent. There is, of course, much more that could be said about this example. The point here is simply to illustrate that there may exist multiple sets of beliefs, each of which is internally coherent, but which are mutually incompatible.

So while coherency is important, it is insufficient to provide certainty. As I mentioned in Belief Optimization, evaluating the coherency of our beliefs in an accurate way is difficult enough, and by practical necessity we cannot independently assess all of the relevent evidence; we must rely on the work of many others. That, however, introduces beliefs about the trustworthiness of those others into the belief system.

Now, I do think that Christian belief systems are less coherent than some atheistic or even deistic alternatives. Possibly I am wrong -- anybody could be. I have previously spent a fair amount of time advancing the claim that eternal consequences for decisions made under such uncertain conditions is unjust. Rather than get into that again, let me raise several other considerations. First, could the uncertainty be reduced? Is there anything that God could do (if he exists) to reduce the uncertainty? Some Christians like to claim that those who do not believe would reject him no matter what he did, but that is presumptuous. It may very well be true of some people, but there are those of us who have taken a measured look at the available evidence and attempted at least to make a reasoned decision. We are willing to consider new evidence, but what we have so far is insufficient. Any explanation that Ernie offers for why eternal damnation would be considered just needs to explain as well why God is unwilling to make our situation clearer and more certain.

Second, while what I have said so far in this post may make it seem like there is very little hope for finding the truth, practically speaking we do reasonably well. For some harder problems, with enough eyes looking and enough time, we often arrive at answers that are "good enough". Sometimes we are wrong; sometimes people die for those mistakes, but we can then learn from those mistakes. Sometimes what we learn is wrong too, but slowly we make progress. Medicines are developed with unforeseen side effects, but eventually the pattern is recognized and medical practice changes. As sad as we may be over the mistakes we make, and as horrible as the consequences sometimes are, we can make progress. Compare this, however, with heaven and hell: we have no observational capabilities that can help us with that. We have no way of verifying the existence, let alone the character, of any kind of afterlife, and certainly no way of correlating any eventual disposition with actions or beliefs during this life. Yet we are told the consequences are beyond anything we can imagine (in duration if nothing else). And this same lack of verifiable progress plagues various aspects of theology.

Whether or not it happens to fall within the context of this dialog, my thought is to explain more fully the variety of reasons that I find Christian beliefs untenable, what preferable alternatives exist, and why they are preferable. I think we have had too many claims and too little explanation and support of those claims.