I very much appreciate the genial tone of Professor Johnson's letter and invite further conversation elsewhere, and I thank the editor for space here to elaborate on aspects of my position about which Johnson raises good questions.
The comments providing context for his use of the term "methodological atheism" are especially helpful. I have not read Reason in the Balance and did not know that this term was (apparently) first used by Nancey Murphy — a very interesting point. I had known of his use of the term from friends who are close to the "Intelligent Design" (ID) movement.
It is also helpful to see how Johnson distinguishes between "naturalism" and "atheism" and that he views the former as more dangerous to religion than the latter. I would say myself that atheism is a religious interpretation of the world, based on an extrapolation of methodological naturalism into ontological naturalism — an extrapolation that is certainly not necessary for doing good science (as various historical examples would illustrate well) and that begs the question of whether truth can be attained apart from methodological naturalism. Science is an "as if" story about natural phenomena that assumes, rather than demonstrates, that all things happen "as if" they had only natural causes.
I agree with Johnson that Ockham's razor would be applied by many to cut away any explanations of any phenomena (whether or not they had only natural causes) that appeal to agents or causes beyond those recognized as natural, but I would call for us to recognize (again) that Ockham's razor is itself a methodological principle that originates outside of science per se; that is, it regulates what counts as a proper "as if" story and cannot be regarded as infallible. Who are we to say, really, what causes could or could not be producing all the events in the whole universe? Nor do all practitioners of a given science agree what is the "simplest" explanation, even without considering agents or causes beyond the natural. And who or what determines when explanations are "multiplied beyond necessity," to cite another form of the principle? Necessary for what, and to whom? To state categorically, for all purposes, that religious explanations of events go beyond necessity is to beg the question of whether religion itself is necessary, and for what purposes.
Questions such as these cannot be decided by "science" which is one important reason why the founders of the Royal Society tried to establish a forum free from discussions of religion and politics — a goal they found impossible to implement in practice. There are legitimate truth questions that science cannot answer with "as if" stories constructed according to its own rules. Indeed, the very reason why science has attained such a high level of prestige in our culture is that it has restricted its inquiry, or tried to restrict its inquiry, to questions for which "as if" stories can be constructed -stories that are capable of gaining a consensus within the scientific community. Might it not be the case that people disagree — that is, they lack a consensus — about moral and religious beliefs precisely because they are more important than scientific beliefs, since they deal more openly and directly with values? Such a question cannot be answered apart from a direct appeal to those same values, and thus defies a response that could be called "objective" in the usual sense. I believe in the primacy of "values" over "knowledge," which is why I call for genuine pluralism in publicly funded schools, whether this is achieved by vouchers or by wholly re-imagining what counts as a "public" school. The problem is not that we rule God out of science classes, but that we rule God out of schools entirely, disenfranchising a large part of the citizenry; this is a political issue not a scientific one.
This is not to say that I reject Johnson's belief that evidence for theism can be found in the world — even within the world that science has constructed for us according to its own rules — but I do not always look for it in the same places. As William Whewell stated in a passage quoted by Darwin opposite the title page of the first edition of the Origin of Species, "we can perceive that events are brought about not by insulated interpositions of Divine power, exerted in each particular case, but by the establishment of general laws." Like Aristotle, I believe that meanings and mechanisms are both legitimate, complementary, even necessary parts of explanations; I do not accept the false dichotomy between them erected by post-Cartesians. The fact that human beings come from fertilized ova does not mean that we are not, as individual persons, made in the image of God.
*I find evidence of purpose in the astonishing fact, pregnant with meaning, that a deep and often subtle order exists and can be found by rational creatures — in the fact that methodological naturalism is so fruitful, rather than in efforts to demonstrate the inadequacy of methodological naturalism to account for certain natural phenomena.* This fact about the intelligibility of the world is hardly necessary for our evolutionary survival and raises profound questions about why this should be so. Such questions are meta-scientific in nature and have often been asked by great scientists who do not share a common religious orientation. I also see evidence for theism in various anthropic phenomena discovered by cosmology; in the persistent human belief in a meaning for existence that goes beyond our own time and place; in the equally persistent belief in "right" and "wrong" as moral categories compared to considering "good" and "bad" simply as attributes of things that happen; and even in aspects of the biological world, such as the progressive development on this planet of an extraordinarily diverse and interrelated system of organisms, which in some respects mirrors (in my view) the Trinity itself.
Neither Zeus nor Santa Claus represents a serious answer to questions of this type, but many would say that God does. I count myself among them.