I think part of my concern about possible dogmatism at the heart of atheism comes down a sense that the natural sciences, whatever their evident power, are necessarily limited in scope. The empirical, quantitative methods of the sciences simply cannot tell us or explain everything that is interesting about the world. To the extent prominent atheists like Dawkins assume the question of God's existence or non-existence can definitively be settled by the natural sciences alone, they seem to have fallen into the dogmatic ideology of scientism.
I have been particularly wary of those who argue that every field of human inquiry is ultimately reducible to or "consilient" with physics. E.O. Wilson in particular seems to propose a solution (I almost wrote "final solution") to the "two cultures" problem that amounts to a wholesale colonization of the social sciences and humanities (one "culture", in C.P. Snow's account) by the physical sciences (the other "culture").
In reaction against efforts like Wilson's, I have been drawn to Stephen Jay Gould's notion that science and religion (into which he lumps ethics and other philosophical concerns) constitute separate, "non-overlapping magisteria." The simplified version goes like this: science asks "what?" and "how?" questions about the world, religion (alongside philosophy) asks "why?" (and perhaps "what now?") questions; as long as the two do not intrude on one another's territory, everything will be fine.
I've been aware that this distinction is troublesome from a number of perspectives. Adherents of religion are unlikely to be willing to cede the field in all matters of "what?" and "how?", and scientific discoveries can have a profound impact on our framing of "why?" questions.
What I have discovered from reading Dennett, though, is that I have been conflating two separate questions. On the one hand is the question of whether the natural sciences constitute the only legitimate form of critical inquiry. On the other hand is the question of whether theistic religion is still tenable in the face of critical inquiry.
My main concern has been with the first question: I have been eager to leave open some intellectual space in which the humanities and social sciences are recognized as legitimate forms of critical inquiry that do not derive their legitimacy from the natural sciences. Dennett uses the old German terms Naturwissenschaften and Geisteswissenschaften for these two broad types of inquiry.
I have not been so much concerned with the second question. It seems simply obvious to me that theistic religious belief is untenable unless truly extraordinary accommodation is made for it, even up to Luther's admonition that we "tear the eyes out of our reason." I will not follow Luther's advice, of course, and my current reading in the literature of the new atheism is spurring me to make even less accommodation to theistic claims in public discourse. My point here, though, is that the two questions really are independent of one another. I'll focus on the first question for a while longer and return to the question of theistic belief later.
I've elsewhere sketched out what I think is a fruitful way of grasping the relationship between the natural sciences and the "human sciences" (if I may be permitted to use "sciences" as a general term for critical intersubjective inquiry), drawing from Kant's "two standpoints": the "spectator perspective" that considers the domain of human experience "from the outside" (e.g., the natural sciences) and the "agent perspective" that considers human consciousness "from the inside" (e.g., phenomenology, ethics). These are two legitimate perspectives on the world of common experience, each of which is authoritative in its own domain, and neither of which can be wholly absorbed into the other.
Here's an illustration (from my "Darwinian Humanism" paper):
A room full of people is a physical space occupied by natural objects that are demonstrably subject to natural laws, and it is also at the same time a moral space in which free moral agents can negotiate the terms of their relations to one another and engage in inquiry and deliberation about what is good and what is right. It is possible to hold these perspectives at the same time because neither on its own can capture the whole truth of what a room full of people really is.
I would clarify that I no longer think of these perspectives as "non-overlapping" - nor even as "magisteria". If anything, the natural sciences and the human sciences are just two sets of tools for inquiry into the world of common experience.
As for the actual practice of inquiry, there is a great deal to be gained from interdisciplinary work across the cultural divide. When we are grappling with complex human problems, we should use the full array of available tools.
My own field of research is environmental ethics and policy, with a particular emphasis on metropolitan growth in the United States. I'm currently collaborating with an economist in a study of the limits of ethics in relation to a particular policy decision. We're looking at those limits both "from the inside" as a matter of lived experience and "from the outside" as a function of various kinds of systems (markets, sociotechnical ensembles, etc.). It is a remarkably fruitful interaction.
I would also add that I have been influenced by Dennett's image (in Freedom Evolves and elsewhere) of Darwinian evolution as a "universal acid", which cannot be contained as it spreads through human inquiry and human experience, transforming everything. My point in the "Darwinian Humanism" paper is precisely that the humanities are transformed as well - though I would insist that this transformation is not a simple dissolution. The sciences may legitimately try to explain consciousness, freedom, and so on, but they do not thereby explain them away. Even scientists still inhabit a meaning-rich world of intersubjective experience in which they freely decide what to do, however they may squint at that world in doing their research.
In the paper, I offer a brief sketch of a phenomenological ("from the inside") approach to understanding scientific inquiry: we start with and remain within the actor perspective, pushing and pulling on the surrounding world, and gradually begin to model the "hinges" and "pivots" (Merleau-Ponty's term) back there in the depths of the world in which we live and act; the natural sciences are one particular way of systematizing that pushing and pulling, using ever more sophisticated tools. With the Darwinian revolution,
we have progressed so far in this systematization that we can begin to model the hidden armature of the actor perspective itself as a product of the evolutionary history of these bodies that we are. We have, in a sense, managed to turn ourselves inside-out. Now that we are here, investigating the natural origins of what we are, we have begun to worry about the creeping advance of mechanistic explanation that seems poised to explain away our freedom and our dignity. And yet, we remain rooted in freedom, and the systematization of knowledge remains one of our projects as incarnate subjects, another expression of our agency.
If there can be any such thing as Darwinian humanism, then, it is not so much a coherent worldview as it is an acknowledgment of an unavoidable ambiguity at the heart of human moral experience: we are somehow able to experience ourselves as fully free and fully natural at the same time.
Now, getting back to atheism, it seems to me that, in the wake of the Darwinian revolution, "the God hypothesis" in its various forms becomes increasingly untenable from the inside as well as from the outside. Just ask Nietzsche.
Of course, it may be that the God hypothesis was always untenable in the face of robust critical inquiry, as its advocates are unable to offer evidence and argument that would be compelling to anyone who does not already believe. The Darwinian revolution simply brings that inherent flaw in theism to its crisis.