bd-from-kg

Ojuice5001:

This whole argument is hopelessly muddled from beginning to end.

You say:

This is a useless distinction, because we can never know what will always be the case; we can only know what is the case now. Thus, at present there are events that we cannot fully explain naturalistically, and it is very likely that this will be true for at least the next few thousand years. But how can we possibly know whether any of these are in “Category C” – i.e., that we will never be able to explain them naturalistically?

This paragraph is logically incoherent. Supernatural events by definition cannot have “characteristics” of the kind required for this paragraph to make sense. And supernatural causes (if this concept even makes sense) would have characteristics indistinguishable from natural causes. More on this later.

This is flat-out false. Haven’t you heard of chaos theory?

Wrong. There cannot, in principle, be such a thing as a “supernatural law”. This becomes clear when we understand what is meant by a “natural law”.

Suppose that after observing the universe for a while, we notice certain patterns and regularities. For example, we may notice that whenever a significant amount pure sodium is dropped into a body of water (a bucket, lake, etc.) the result is an explosion. We check this by conducting a number of controlled experiments, but the result is always the same: sodium + water = explosion. We are then justified in formulating a natural law: “If sodium is combined with water, the result will be an explosion.” (Of course, if we are real scientists, we won’t be satisfied with this; we’ll want to know “why” we always get an explosion when we combine sodium with water. But we’ll never get an ultimate “why”; all we’ll get, at most, is that this is a special case of more general regularities.)

But wait! How do we know that this is a natural law? Maybe it’s really a supernatural law! In other words, maybe we always get an explosion because whenever sodium combines with water God makes an explosion!

The problem with this idea is that, scientifically, these are identical hypotheses. There is simply no difference, scientifically, between saying that you always get an explosion under these conditions and saying that God always creates an explosion under these conditions. The two hypotheses yield exactly the same predictions in all cases, which is to say that operationally there is no difference between them. So the so-called “supernatural law” that God always creates an explosion when sodium is combined with water is in fact a natural law; it is the same natural law that we stated originally; the only difference is that it is expressed differently.

The same thing holds in all cases. For any conceivable pattern or regularity that you might find, you can state a corresponding natural law. Or, if you prefer, you can formulate a corresponding “supernatural law”. So long as the two describe the exact same regularity or pattern, they are scientifically identical. Thus it is meaningless to call one a “supernatural law” and the other a “natural law”; they are both natural laws.

But suppose that the supernatural law does not describe exactly the same pattern or regularity as the “natural law” it is proposed as an alternative to? Well, in that case we can consider a modification of the proposed natural law that is operationally identical to it, and then run the appropriate tests to determine whether it gives better predictions than the original natural law. If so, we adopt it. Or, if you prefer, we can adopt your “supernatural” law. Scientifically, it makes no difference. What matters is whether the pattern or regularity in question really holds uniformly. If it does, it doesn’t matter (scientifically speaking) how it is described.

Now let’s consider the question of why we would call one way of describing a pattern or regularity as a “supernatural law” in the first place. Presumably it is because somewhere in the description there is a mention of a hypothetical supernatural agent. Such references can always be removed. This will perhaps be clearer if we consider the motion of the planets. We might propose to replace the classical account of why the planets move the way they do (i.e., an inverse-square gravitational force) with the explanation that God keeps them in their orbits. But we still have to account for why they move exactly as if they were under the influence of an inverse-square force, so we will have to add that God causes them to move in just the way that an inverse-square force would. But now we can translate this into a law with no mention of a supernatural agent by replacing “God” with “X”. We now have “X causes the planets to move in just the way that an inverse-square force would”. But since X is simply an unknown factor, we might as well give it a name. How about “gravity”? (We must understand here that gravity is not something that we know anything about, but simply a convenient name for the unknown factor “X”.) We now have “Gravity causes the planets to move in just the way that an inverse-square force would.” But this is awkward; we might just as well say, “the planets move the way they do because they are under the influence of gravity” and think of gravity as an inverse-square force. It is understood that this “gravity” is just a name for we-know-not-what; all that we know about it is that it causes the planets to move the way they do. For all that we know or care as scientists, “gravity” might be the name of something with “supernatural” properties; it might be God. But all that we know about it is that it affects the planets the same way that an inverse-square force would, and so it is convenient to think of it as an inverse-square force.

Thus we finally have this result: it is possible to express an observed pattern or regularity in terms of hypothetical supernatural agents or forces, but it is always possible to “translate out” all such references. And the result, so far, has always been to obtain a simpler description of the pattern or regularity. Indeed, it seems clear that this must always be the case, because the hypothesis that an agent or cause has supernatural properties is always superfluous. So, contrary to your conclusion that “supernatural explanations are the best way to go” in some cases, we now see that they are never the best way to go. They just result in unnecessarily complicated descriptions of the patterns or regularities involved, and thus always violate Occam’s Razor.

One last point. You say that “true randomness is utterly unlike the rest of our knowledge”. But then, you’ve already segregated out “chance events” – i.e., ones in which true randomness plays a role - so by definition true randomness must be utterly unlike anything that we know about anything else. This is a tautology masquerading as an argument.

And anyway, it isn’t clear that there is any “true” randomness in nature. It’s true that the Copenhagen interpretation of quantum mechanics postulates an element of true randomness, but at least two other interpretations – the transactional interpretation and the many-worlds interpretation – are completely deterministic. There is no other element of “true randomness” in the “standard model”, or for that matter in any physical theory that I know of that is currently “in play”.

[ August 18, 2002: Message edited by: bd-from-kg ]</p>

hinduwoman

let me get this straight; are you saying that whatever happens by chance is guided by supernatural powers?

Jobar

"In that case, there would be two categories of events. One category of events would be the events that can be fully understood by confining your attention to natural factors. The other category of events--let's call it category C--would always remain partly unpredictable by science. "

Ojuice, there are lots of events and systems which science cannot precisely predict- the weather being one example. This, however, is simply because we can't precisely specify the initial conditions, and the enormous number of variables mean that no computer could handle the equations. As bd says, quantum systems are predictable only in a statistical sense, and no way will ever be found to, say, exactly specify the time when a uranium atom will decay. However, the statistical predictions we derive from quantum mechanics are incredibly precise; I can't see how you can call quantum uncertainty in any way 'supernatural'.

In short- I don't know of any events which really fit into your category C. So 'supernatural' seems to be a null concept.

Ojuice5001

Yes. As I detail in my survey of the alternatives, I think that this concept allows the world to have room for the supernatual, and no other (except deism) does.

One-eyed Jack

bd-from-kd, you covered virtually everything I had in mind to write.

Ojuice5001, you are confusing "random" with "complex beyond analysis" and "chaotic." Neither of the last two situations are random.

A God who worked only through influencing random events could only work with quantum probabilities (in the Copenhagen interpretation) and even then He would have to precisely duplicate the quantum probabilities we measure because it seems that experimental evidence indicates there are no "hidden variables" behind the quantum. In other words, the evidence suggests God cannot manipulate quantum events in a purposeful way without violating physics: a miracle.