Five days ago I posted an account of Michael Shermer’s talk at the atheist meeting in Mexico City, noting that he denied the possibility of the supernatural and therefore ruled out the existence of God on first principles. I disagreed with him, arguing that the presence of a divine being, one able to either obviate or manipulate the laws of nature, was at least a hypothetical possibility. The proper scientific attitude, I claimed, is to say that based on a lack of evidence for a god (and the presence of evidence arguing against the existence of an intrusive and beneficent being), we can therefore ignore a god hypothesis.
There was a time, however, when the hypothesis of God was not incoherent, but viable. I refer not only to pre-Newtonian physics, but also to biology before Darwin: creation by a supernatural being was the only going theory before the 19th century, but Darwin put paid to that. It is the juxtaposition of scientific evidence with a creationist, supernatural theory making unsubstantiated claims that made Darwin’s Origin so convincing.
Michael emailed me promising to respond to my post and to defend his original view that there can be no such thing as the supernatural. He has now done so in his column over at Skepticblog, in a post called “God, ET, and the supernatural“.
Shermer’s argument is simple: we can’t distinguish between a supernatural being and an advanced civilization of, say, extraterrestrials that could perform all the “signs and wonders” that would convince most of us that God exists. As he notes,
My argument is that the most any natural science could ever discover in the way of a deity would be a natural intelligence sufficiently advanced to be god-like but still within the realm of the natural world. As I wrote in Scientific American:
“God is typically described by Western religions as omniscient and omnipotent. Since we are far from the mark on these traits, how could we possibly distinguish a God who has them absolutely, from an ETI [extraterrestrial intelligence] who has them in relatively (to us) copious amounts? Thus, we would be unable to distinguish between absolute and relative omniscience and omnipotence. But if God were only relatively more knowing and powerful than us, then by definition it would be an ETI!”
Well, yes, we wouldn’t know whether a divine being was absolutely omniscient and omnipotent, or relatively more omniscient or omnipotent than us. But if the degree of, say, omnipotence and omniscience is sufficiently large (i.e, any miracle can be worked, all things can be foretold), then I think we can say provisionally that there is a God. I’ve previously described the kind of evidence that I’d provisionally accept for a divine being, including messages written in our DNA or in a pattern of stars, the reappearance of Jesus on earth in a way that is well documented and convincing to scientists, along with the ability of this returned Jesus to do things like heal amputees. Alternatively, maybe only the prayers of Catholics get answered, and the prayers of Muslims, Jews, and other Christians, don’t.
Yes, maybe aliens could do that, and maybe it would be an alien trick to imitate Jesus (combined with an advanced technology that could regrow limbs), but so what? I see no problem with provisionally calling such a being “God”—particularly if it comports with traditional religious belief—until proven otherwise. What I can say is “this looks like God, but we should try to find out more. In the meantime, I’ll provisionally accept it.” That, of course, depends on there being a plethora of evidence. As we all know, there isn’t.
My sharpest disagreement with Shermer is his denial that there could even be the possibility of a divine being. Even if aliens could imitate one, to me that doesn’t rule out the possibility of a god, though we might be unable to distinguish that god from advanced ETIs. Where Shermer and I differ, then, is in how we regard “evidence” for a God. At some point I would just say, “Okay, I’ll tentatively say there’s a God,” while Shermer would always say, “We can’t tell if it’s God or an ETI,” even if we have no independent evidence for ETIs. He considers his hypothesis more parsimonious because he believes evolution could produce advanced beings with “godlike” powers, given the immensity of the universe and the immensity of time. Well, maybe, but the universe is only 14 billion years old and habitable planets are much younger than that. Evolution can only do so much in a few billion years.
Shermer’s whole argument rests on the fallacy that scientists are committed to methodological naturalism: we can accept only natural explanations for natural phenomena:
On the matter of the supernatural, Jerry Coyne continues in his blog:
“As always, I find the natural/supernatural distinction confusing, and see that it is possible in principle for some divine being who operates outside the laws of physics to exist. To say there is no possibility of such a thing is an essentially unscientific claim, since there is nothing that science can rule out on first principles. We rule out things based on evidence and experience, that is, we consider the possibilities of gods extremely unlikely since we have no good evidence for them. But it is close-minded to say that nothing would convince us otherwise.”
I disagree. It is simply a matter of what philosophers of science call methodological naturalism, or the process of employing only natural explanations for natural phenomena. Science operates in the natural, not the supernatural. In fact, I go so far as to say that there is no such thing as the supernatural. There is just the natural and mysteries we have yet to explain by natural causes. Invoking such words as “supernatural” (and, in other realms, the “paranormal”) just provides a linguistic place-holder until we find natural causes (or we do not find them and discontinue the search out of lack of interest).
. . . A supernatural entity or force (something like the God of Abraham) that exists outside of nature is, by definition, unknowable to science. By contrast, if a supernatural being reaches into our natural world in order to act on it, He must stir the particles in some way (to, say, answer prayers for healing a cancerous tumor by reconfiguring the DNA of the cancerous cells, or to help one nation win a war over another by redirecting bullets and bombs, or to aid one football team defeat another in the Superbowl by deflecting a touchdown pass), and that action must in principle be measurable by science. If it is not measurable even in principle, then it is not knowable by science.
Like Michael, I think the word “supernatural” is a bit ambiguous, for “supernatural” beings—one who can obviate or manipulate the laws of physics—can create natural phenomena. The being might not be demonstrable, but the actions of that being might well be. In that sense there can be natural evidence for a supernatural god. We can’t see electrons, either, but we can see their actions, and hence infer that they exist.
Shermer thinks that a “supernatural” being must do things that obey the laws of physics, but I claim otherwise. (And how, by the way, do we know when the laws of physics or chemistry have been violated? After all, they have been reliable generalizations, but maybe we don’t know everything about them.) A real god could obviate the laws of physics in such a way that we could never understand godly phenomena (like answered prayers) except as something inexplicable under current knowledge. At that point Shermer defaults to aliens, while I start thinking about the supernatural.
I don’t see science as committed to methodological naturalism—at least in terms of accepting only natural explanations for natural phenomena. Science is committed to a) finding out what phenomena are real, and b) coming up with the best explanations for those real, natural phenomena. Methodological naturalism is not an a priori commitment, but a strategy that has repeatedly worked in science, and so has been adopted by all working scientists.
As for me, I am committed only to finding out what phenomena really occur, and then making a hypothesis to explain them, whether that hypothesis be “supernatural” or not. In principle we could demonstrate ESP or telekinesis, both of which violate the laws of physics, and my conclusion would be, for the former, “some people can read the thoughts of others at a distance, though I don’t know how that is done.” If only Christian prayers were answered, and Jesus appeared doing miracles left and right, documented by all kinds of evidence, I would say, “It looks as if some entity that comports with the Christian God is working ‘miracles,’ though I don’t know how she does it.”
My view on Shermer’s naturalistic “commitment” was expressed by one of his commenters:
Explicit Atheist says:
Michael Shermer argues “It is simply a matter of what philosophers of science call methodological naturalism, or the process of employing only natural explanations for natural phenomena.”
I completely disagree. Science is about discovering what is true about how our universe works. If divine revelation given to those who worship a particular deity in a particular way was the method that worked then science would adopt methodological supernaturalism and scientists would be people who devote themselves to obtaining revelations about how the world works by worshipping that deity that way. Mr. Coyne, Stenger, and Dawkins are correct, science a- priori presumes nothing and rules nothing in or out. Science is completely pragmatic and will adopt any methodologies and any conclusions that are successful. Success is the only criteria that defines what is scientific and what is not, both up-front with methodologies and down-back with conclusions.
In the end, the difference between Shermer and I comes down to this: if evidence were really pervasive for an immensely knowledgeable and powerful being, I would tentatively accept God, while Shermer would tentatively accept an ETI that that works in unknown (but natural) ways. He is unwilling to say that there can be anything other than the natural world; I claim that this is a good working hypothesis but one that can never be verified with absolute certainty.
Science can never prove anything. If you accept that, then we can never absolutely prove the absence of a “supernatural” god—or the presence of one. We can only find evidence that supports or weakens a given hypothesis. There is not an iota of evidence for The God Hypothesis, but I claim that there could be.