More hints about the concept of a “scientific question” from the Charles Simonyi Professor of the Public Understanding of Science …
In The God Delusion, Dawkins gives a couple of hints about what constitutes a scientific hypothesis.
A scientific hypothesis is an idea that is either true or false.
Dawkins is more favorable towards TAP style agnosticism, than PAP style agnosticism:
TAP, or Temporary Agnosticism in Practice, is the legitimate fence-sitting where there really is a definite answer, one way or the other… (TGD, p.47, emphasis added)
Dawkins connects the truth-or-falsehood of the idea that “God exists” with the allegedly scientific status of the question:
Either he [God] exists or he doesn’t. It is a scientific question… (TGD, p.48)
Dawkins says something similar in relation to questions about miracles:
…this [the question about whether Jesus’ mother was a virgin at the time of his birth] is still a strictly scientific question with a definite answer in principle: yes or no. (TGD, p.59, emphasis added)
An idea that can be true or false is what philosophers call a proposition. Many philosophers hold that moral and normative judgments are neither true nor false, thus such judgments would not be considered to be propositions. Dawkins’ condition here, would, on this view of the nature of moral judgments, exclude such judgments from being scientific hypotheses, and this would in turn exclude the questions for which those judgments are answers, from being scientific questions.
However, not much else is excluded by this condition. Many historical and mathematical claims and hypotheses, for example, are also either true or false. So this condition only works as a necessary condition for an idea to be a scientific hypothesis, not as a sufficient condition.
Another closely related hint concerns the existence of conflicting alternative ideas.
A scientific hypothesis is a proposed idea that has implications that clearly conflict with the implications of some other alternative idea.
Dawkins thinks that the contrast between atheism and theism is relevant to considering the God hypothesis to be a scientific hypothesis:
…a universe with a supernaturally intelligent creator is a very different kind of universe from one without. (TGD, p.58)
…[the idea of a] non-interventionist…God…is still…a scientific hypothesis. I return to the point: a universe in which we are alone except for other slowly evolved intelligences is a very different universe from one with an original guiding agent whose intelligent design is responsible for its very existence. (TGD, p.61)
Because atheism and theism have very different and conflicting implications, this seems to be a reason, for Dawkins, to consider the belief that “God exists” to be a scientific hypothesis.
Again, this works as a necessary condition, but not as a sufficient condition, for claims and hypotheses in history, mathematics, and logic also can be contrasted with alternative ideas that have very different and conflicting implications.
(1) Evidence should be applicable to answering the question.
(2) The question should be answerable in principle, if not in practice.
(3) Some proposed answers to the question should be either true or false.
(4) One proposed answer to the question should have different and conflicting implications in comparison to some other proposed answer.
Taken individually, each necessary condition encompasses a broad scope of questions that go beyond the narrower scope of questions encompassed by the concept of a “scientific question”.
What if we take all of these conditions together to be a sufficient condition for something being a scientific question? This won’t work, because the same set of counterexamples satisfy all four necessary conditions; many historical, mathematical, and logical hypotheses would satisfy all of these conditions.
There is one more hint from Dawkins that seems to have a bit more substance to it. He sometimes refers to “the scientific method” or “scientific methods” when discussing the nature of religious beliefs. Dawkins quotes Alister McGrath about Huxley (a PAP agnostic, according to Dawkins):
“…Huxley declared that the God question could not be settled on the basis of the scientific method.” (TGD, p.54)
Dawkins also quotes McGrath’s quote of Stephen Gould:
“…science simply cannot (by its legitimate methods) adjudicate the issue of God’s possible superintendence of nature.” (TGD, p.55)
Dawkins himself uses the phrase “scientific methods” when discussing how miracle claims should be evaluated:
There is an answer to every such question, whether or not we can discover it in practice, and it is a strictly scientific answer. The methods we should use to settle the matter, in the unlikely event that relevant evidence ever became available, would be purely and entirely scientific methods. (TGD, p.59, emphasis added)
Dawkins implies that a question is a scientific question if the question can (in principle) be settled by the use of methods that are scientific methods.
Here we finally have what appears to be a sufficient condition, and a plausible one at that. However, although this may truly be a sufficient condition for something to be a scientific question, it is not very helpful or useful as it stands.
We need to have some clear explanation or specification of what methods constitute legitimate scientific methods, and what methods do not. Furthermore, additional guidance might also be required in order for us to be able to determine whether a particular question can “in principle” be settled by the use of such methods.