Tuesday, November 22, 2011

Thinking, and non-thinking, atheists

It's understandable that atheists should have a problem with the "supernatural" but it's remarkable how had some of them seem to find it to think straight when this is mentioned.

The Faraday Institute publicised a blog post in Scientific American which claims:
"To the extent we can be certain about anything, we can rest assured that all supernatural claims are false"
This is such arrogant nonsense that it is hard to know where to begin.  Not only is no reasonable argument presented for this claim, it is very hard to see how a reasonable argument could be presented for this claim. Certainly there is no conceivable set of scientific observations that could substantiate it. Even if there were a known set of deterministic scientific laws whose predictions agreed completely with every experiment that had ever been performed (which is certainly not the case) that would not and could not demonstrate that “all supernatural claims are false”.

But in a gushing eulogy for Lisa Randall's Knocking on Heaven's Door  in Science Michael Shermer comes out with this:
if divine providence were on the offing, “it is inconceivable from a scientific perspective that God could continue to intervene without introducing some material trace of his actions.” In other words, if God did act in the world scientists would want to know how he did it. “Did He apply a force or transfer energy?” Randall asks rhetorically. “Is God manipulating electrical processes in our brains? … On a larger level, if God gives purpose to the universe, how does He apply His will?” Inquiring minds want to know. Religion has no answer. I know because I have asked many times.
 He (and Randall?) seem to be "arguing" that unless we have a clear scientific explanation of how X happens then X cannot occur.  Now firstly if we are talking about God interacting with the Universe there is no reason at all to demand that this occurs through "normal" physical processes. The relationship of the Ulimate Creator to the universe is analagous in many respects to that of a programmer and a simulation.  Even if in the simulation a physical law applies programmers can and do interact with the simulation in ways that are completely different from the internal physics.  This is so blindingly obvious in this age of CGI that it is amazing that people can overlook the point with a straight face.  Even if Shermer doesn't do much/any real science Randall surely runs simulations and certainly knows people who does.

But secondly if the doctrine were to be taken seriously it would rule out doing any real scientific research.  The whole point of research is that we look for things that do occur for which we don't have a current scientific explanation and try to find one.

This brings me to Julian Baggini (who I rather like) and his attempt at "Articles of 21-st century faith".  He also wants to rule out:
"claims about the nature and origin of the natural universe...belief that any supernatural events have occurred here on Earth, including miracles that bend or break natural laws, the resurrection of the dead, or visits by gods or angelic messengers... the thoughts of a divine or supernatural mind that exists independently of humanity"  
and asserts blythely that
"if a religion makes a claim that is incompatible with our best science, the scientific claim, not the religious one, should prevail"
There are many confusions here but one of the most fundamental seems to me to the implicit assumptions about the nature of science. It assumes that there is something called "our best science" which is an essentially complete and definitive description of everything set of 'facts' {I've revised this following Julian's comment} about "the nature and origin of the natural universe". Now let's assume for a moment that there is such a thing as the complete physical laws of the universe (this is a metaphysical assumption historically rooted in theism and could be false, BTW).  Call this set L*. Clearly we don't have L* but some approximation Ln say, and it is very hard to know how we could ever verify that we had L*. But even if we had L* and knew that we did, L* cannot possibly explain the origin of L* and we would have to appeal to metaphysical/theological principles to do so.

Furthermore to return to the reality of Ln, there is hot debate and rightly so about whether some of the ideas of our "best science" are right at all.  Most cosmologists currently believe in something like string theory and our "best science" certainly contains multiverses etc.. although much of this is driven, as Martin Rees admits, by a desire to avoid the otherwise compelling arguments for creation.  But the empirical evidence for this is inadequate and it is perfectly reasonable to put forward alternative hypotheses.  The motivation for alternative hypothesis is basically irrelevant to their scientific validity.  Attempts to discredit "Big Bang" on atheological grounds should give people like Julian some pause for thought.

Science is not fundamentally a collection of 'facts' but a set of theories and observations. Some of the observations turn out to be wrong (see eg this paper in Science) and the relationship between theory and observation is highly complex and intertwined. I am not suggesting that the concept of scientific fact is vacuous or that anything goes in the domain of claims made by religions. But religious claims and scientfic claims are almost invariably different kinds of claim and notions of "incompatability with our best science" are highly problematic. 

Indeed it is interesting to reflect that in many respects 20th century science has moved quite a long way in a "Christian" direction.  As far as all empirical science knows, there was a big bang which is suspiciously like creation, but in addition it is now quite clear that "dead" people sometimes come back to life and that in extreme cases the "placebo" effect is very strong indeed, with very strong interactions between the nervous and immune systems.  Indeed I don't think any of the healing miracles in the Gospels can now be considered as impossible.  And of course, as John Polkinghorne and others have pointed out, we really don't seem to live in a mechanical, deterministic universe.

3 comments:

Julian Baggini said...

Nicholas, you say: 'There are many confusions here but one of the most fundamental seems to me to the implicit assumptions about the nature of science. It assumes that there is something called "our best science" which is an essentially complete and definitive description of everything about "the nature and origin of the natural universe".'

What seems to you just isn't the case. I make no such assumption and so all that follows is besides my point. It's you who makes the false assumption about what I assume!

starcourse said...

Julian:
Thanks. I've amended the post slightly. But it still seems to me that the "nature and origins" of the universe is in the sense described beyond the domain of science, and that the concept of "incompatible with our best science" is quite problematic.

Iain said...

Thank you for the post. The simulation idea has often occurred to me (though I'm not sure it's the best model for God interacting with the universe).

Suppose I write a computer simulation of a number of bodies interacting under gravity (e.g. a solar system, or binary star). The program outputs, at each time step the position and velocity of each of the bodies. Suppose further, I put a break point in the simulation and run it under the debugger (something I do every day in my normal work). Then when I hit the breakpoint, I could, for example alter the contents of the memory location that is used for the gravitational constant. I restart the simulation from that point, and the behaviour changes accordingly.

Now suppose someone at a later date examines the output from the program - makes an animation or whatever. They will observe that the gravitational constant changed at some point, for which they will have no scientific explanation, though they could otherwise predict what was going to happen while it remained constant. The fact of the one-off intervention from outside the system does not preclude an observer doing science on the rest of it.

If, however, I arbitrarily change the gravitational constant at every single time step, then the observer of the simulation output will not be able to make any sense of it at all.

So I would say that continual supernatural interventions will preclude science, but it does not follow that _occasional_ supernatural interventions are precluded by science.

 
/* Google Analytics */