|
I've often brought up IP research to contrast to the nonsense pushed by Creationists, particularly Intelligent Design advocates. If done right, IP research can be science becasue it relates two measurables, one a manipulation, an intervention: (1) having a group pray for an experimental group; (2) a measured outcome. If done rigorously as a randomized, double-blinded experiment to test pre-specified hypotheses, an IP experiment could do good science while trying to uncover a causal connection between 1 and 2. True, the specified hypothesis tested is not scientific in that it does not propose a naturalistic mechanism, but science often has reason to test such hypotheses -- e.g.: clinical trials of 'alternative' medical interventions; skeptical debunking of occult claims; and so on. The real scientific issue would occur should IP research show strong results upholding prespecified hypotheses, particularly under replication of studies.
It has not done so. In fact, little has been shown. Until the fertility study, I had not seen any study claiming to have upheld a specific prespecified hypotheses. Instead, weak connections had been reported between IP and some set of factors cobbled together after the fact -- in effect, the researchers 'went fishing' for a hypothesis that could be upheld by the data they had already collected and begun analyzing. That can be fine for proposing hypotheses to be tested in future research, but does not uphold any hypothesis.
So, what would be the issue if an IP-outcome connection appeared to be upheld? The issue, of course, would be what the mechanism was by which IP produced an outcome. The model would be
IP --> black box --> outcome
What's in the black box? Of course, the motivation of most IP researchers would be to claim that God was in the black box, answering prayers. But many other hypotheses could be entertained. One would be that the experimental results were somehow contaminated or that the experimental procedure was otherwise 'busted' -- naturalistic mechanisms. Likewise, there might be mechanisms not now known to science but that might someday be recognized as naturalistic. What if someone had stumbled over the ability to propagate radio waves before the development of enough knowledge of electromagnetism to explain what was happening -- wouldn't sending a message via radio have appeared to be 'magic'? If one is to provide supernaturalistic explanations, my favorite is that IP research is inherently sinful and that the devil answers the prayers in order to promote it. ;-) The question would be how to test competing hypotheses. A proposition that a result was pure chance could be tested by replicating an experiment -- one reason replication is important. An argument that the experimental design failed in some specified way can also be tested by replication with more-stringent replication incorporating stronger controls aimed at the suspected source of failure. But what could be done with a God vs. devil argument?
Actually, the fertility results amuse me because the Catholic Church holds that the fertility treatments are immoral -- if those results were found to hold up under replication, the Pope might argue that it was the work of the devil after all. ;-) But, likewise, those who think fertility intervention is just fine morally would argue that God answered the prayers. How could that disagreement be resolved? That is the crux of the problem noted by Dr. McNally -- but that invalidates the interpretation of the research, not the research itself, as science. So, replicate the fertility IP experiments and see whether consistent positive findings emerge. Then the mechanism can be argued over.
Now, it's another issue whether the government should fund the research. The Constitution establishes separation of church and state, not church and science. It may be that the founding fathers' motivation was to take the state out of debates that could have no natural, so no scientific, resolution; but it is at least as likely that the motivation was to reduce religious fighting using the state and to promote freedom of belief, even if beliefs are unfounded. In that case, could it be that government is prevented from funding IP research if it is motivated either by (1) belief that God does it, or (2) a skeptical wish to debunk IP? Possibly; but even so, that leaves open the avenue of funding IP as a test of 'alternative medicine,' under the rubric that "a lot of people try it and think it works, so let's test it to see whether, for some reason, it really does." I think that government funding can be justified under that rubric, but I'm not sure how much funding should go that way with what priority. There should certainly be some point in which nonresults lead to a decision to abandon such funding.
In the end, my concern is that science be defended from nonscientific thinking, such as Creationsim. In that regard, I would much prefer that those advancing supernaturalistic arguments use the best scientific approaches to test them, rather than trying the end runs that, in particular, Intelligent Design advocates have attempted. In that regard, I will continue to use IP experiments as an example of a correct approach that might be taken.
And what is the meaning of 'natural' and 'supernatural,' anyway. I argue that basically the 'natural' is whatever science has been able to show that it could observe and measure, hopefully manipulate and test, and attempt to explain, and that provides new hypotheses and avenues for research. Simply saying that 'God did it' does not do such things. But at the same time, the exclusion of God from scientific explanation largely since Galileo can be seen as the esult contingent on findings. When Galileo pointed his improved telescope to the heavens, he found the moons of Jupiter rather than a divine throne sitting on a cloud. As an agnostic, I have no problem with a believer who believes in God despite Galileo's, and hundred of years of further oberservers', faliure to sight a divine throne; but as a scientist and an advocate of science, I object to those who attack science's naturalistic methods.
|