Some questions/issues about the relation between science and religion, and in particular Gould's take on this.
0. What's a "magisterium"? And what are the magisteria of religion and science? Gould writes:
"The text of Humani Generis focuses on the magisterium (or teaching authority) of the Church�a word derived not from any concept of majesty or awe but from the different notion of teaching, for magister is Latin for "teacher." We may, I think, adopt this word and concept to express the central point of this essay and the principled resolution of supposed "conflict" or "warfare" between science and religion. No such conflict should exist because each subject has a legitimate magisterium, or domain of teaching authority�and these magisteria do not overlap (the principle that I would like to designate as NOMA, or "nonoverlapping magisteria").
The net of science covers the empirical universe: what is it made of (fact) and why does it work this way (theory). The net of religion extends over questions of moral meaning and value. These two magisteria do not overlap, nor do they encompass all inquiry (consider, for starters, the magisterium of art and the meaning of beauty)."
He apparently says a bit more about what he means by a "magisterium" in his book; Goodenough quotes this definition:
"a domain where one form of teaching holds the appropriate tools for meaningful discourse and resolution."
1. What's the "magisterium" of science?
What kinds of questions can science hope to settle?
My own rough answer to this would be: any empirical question that can be answered at all.
This excludes: (a) questions that are about values rather than facts (what's a good person, what's the right thing to do, etc.); (b) questions that are factual but not empirical, e.g. mathematical questions; (c) questions that may be factual but that we have no way to answer at all (perhaps questions about things that lie outside our event horizon).
It includes: everything else! In particular, to the extent that religions make factual claims (such as: the universe was created by God; the world was created in seven days; in the communion service, the wine changes into blood; etc.), these are in principle testable by science.
Is there any way other than science to answer empirical questions?
My take: basically no, but a more precise answer requires refining the question in a couple of ways.
First, what is it to "answer" a question? Of course we can give answers that aren't based on science: I might offer answers to factual questions that are based on hunches, whims, hearsay, etc. However, I don't think any of these techniques provide answers that we have any reason to accept. The more interesting question is what it takes to give an answer to a question that we are rationally justified in believing.
Second, how broadly should we construe "science"? I don't think there's any way to give justifiable answers to factual questions that doesn't involve empirical evidence, and I think that the standards for what counts as good evidence for a proposition are broadly speaking scientific standards. So if we take "science" broadly enough then I don't think that any factual questions are answerable by any method other than the scientific method.
However, I also think that some good reasons to accept propositions wouldn't count as science in a narrower sense. For instance, I think that it can be rational for an individual to believe something because of the testimony of someone else whom the individual believes (on good grounds) is trustworthy and knowledgeable about the subject.
Another example: it's rational to believe that there's a cup of coffee in front of me on the basis of the evidence of my senses. This an example of believing a factual claim on empirical grounds, but it doesn't seem general enough to count as scientific in the familiar sense.
2. What's the "magisterium" of religion?
Gould seems to think that value questions are the proper domain of religion: "The net of religion extends over questions of moral meaning and value." I disagree. I see no reason to think that insights about ethical or other value questions have any necessary connection with religion. [Cf. Dawkins, The God Delusion, p. 80: "Why not the gardener or the chef?"] In particular, I think that Plato showed convincingly in the Euthyphro that moral judgments do not need a religious foundation.
What does religion involve, anyway?
1. doctrines. To the extent that these claim objective truth, it seems to me that the reasons to believe them (if any) must come from outside religion itself: for instance, from science or philosophy.
2. practices. Religions tend to involve practices such as prayer, singing hymns, and ceremonies such as baptism, confirmation, and communion. To the extent that these are independent of particular doctrines they don't need any theoretical justification; if practioners find them rewarding, that's enough. (On the other hand, the doctrine that prayers are listened to and perhaps answered by a divine being is a factual claim that would require evidence.)
3. values. It may be part of practicing a particular religion to value certain things. (For example, in Catholic theology, faith, hope, and charity or love are regarded as theological virtues, alongside the cardinal virtues of prudence, justice, temperance, and courage.) My own view is that values require justification, and that this cannot come from within religion itself.
As far as I can see, the only way for religion to avoid potential conflicts with science (or philosophy) is by essentially eliminating its doctrinal component altogether. It was common for theologians in the 50s and 60s to do this, or something very close to it. In some of their writings we can see suggestions that look a lot like Gould's doctrine of nonoverlapping magisteria:
"Science can conflict only with science, and faith only with faith; science which remains science cannot conflict with faith which remains faith. . . . The famous struggle between the theory of evolution and the theology of some Christian groups was not a struggle between science and faith, but between a science whose faith deprived man of his humanity and a faith whose expression was distorted by Biblical literalism. It is obvious that a theology which interprets the Biblical story of creation as a scientific description of an event which happened once upon a time interferes with the methodologically controlled scientific work; and that a theory of evolution which interprets man's descendance from older forms of life in a way that removes the infinite, qualitative difference between man and animal is faith and not science." --Paul Tillich, Dynamics of Faith (New York: Harper & Row, 1957), pp. 82-83.
"Theological statements are not a description of 'the highest being' but an analysis of the depths of personal relationships -- or, rather, an analysis of the depths of all experience 'interpreted by love'." --John A. T. Robinson, Honest to God (Philadelphia: Westminster Press, 1963), p. 49.
"Contemporary Christian proclamation is faced with the question whether, when it demands faith from men and women, it expects them to acknowledge this mythical world picture from the past. If this is impossible, it has to face the question whether the New Testament proclamation has a truth that is independent of the mythical world picture, in which case it would be the task of theology to demythologize the Christian proclamation" (Bultmann, p. 3). "We cannot use electric lights and radios and, in the event of illness, avail ourselves of modern medical and clinical means and at the same time believe in the spirit and wonder world of the New Testament" (Bultmann, p. 4). [First quotation taken from Wikipedia article on Demythologization; second from Wikiquote article on Demythologization.]
3. Potential conflicts between the "magisteria."
A somewhat surprising feature of Gould's description of the magisterium of religion is that it leaves out claims about the supernatural, and specifically, claims about the existence of God. Certainly many religious people would regard this as an important component of their religions! (As Goodenough says, religions generally base their views on value and meaning on a "cosmology.") If we add metaphysical claims about God and related matters to the subject matter of religion, then we face the question of whether science can tell us anything about such matters.
Two very different groups have argued that it can. On the one hand, proponents of intelligent design (following a long line of defenders of various versions of the "argument from design") argue that scientific knowledge can provide powerful arguments for the existence of God, by showing that the probability that the universe we know could have arisen without an intelligent designer is vanishingly small. (Dawkins thinks that this argument simply leaves us with an unanswerable question about how to explain God's existence, and so doesn't accomplish anything. This reply doesn't seem as powerful to me as it does to Dawkins. The issue is how to explain the existence and nature of the natural world; the intelligent design theorist thinks we need to invoke the supernatural in order to do this. But the ID proponent may hold that the supernatural realm does not need the same kind of explanation; for example, many theists hold that God exists eternally and necessarily, so that no explanation of how God came to exist is possible or needed.)
On the other hand, atheists sometimes argue that scientific reasoning can show that there probably is no God. Dawkins offers an argument like this. He sets things up by asking, as the ID proponents do, what is the best explanation of apparent design in the universe. One hypothesis is that it was put there by an intelligent designer (God); an alternative hypothesis is that (in the case of biological complexity) it is the result of natural selection. But God is at least as improbable as the complexity he is invoked to explain, making natural selection the better hypothesis.
It surprises me that Dawkins does not address another argument which employs broadly scientific reasoning to argue for atheism, namely the traditional "problem of evil." The world contains lots and lots of suffering which serves no obvious purpose. The argument suggests that the existence of so much suffering would be highly improbable if there is a God, but completely unsurprising if there isn't one. So given the evidence of suffering, the nonexistence of God should be regarded as more probable than the existence of God.
[Notice that we can think of the reasoning here in Bayesian terms. Let G be the proposition that God exists, and let S be the proposition that there is an enormous amount of suffering in the world. P(G|S) = (P(S|G) * P(G)) / (P(G|S) * P(G) + P(~G|S) * P(~G)). P(~G|S) = (P(~S|G) * P(~G)) / (P(G|S) * P(G) + P(~G|S) * P(~G)). The denominator is the same in both cases, so P(G|S) will be higher (or lower) than P(~G|S) just in case P(S|G) * P(G) is higher (or lower) than P(S|~G) * P(~G).
So which posterior probability is greater depends in part on the prior probability we assign to G or ~G. But if we want to set aside the extent to which the posterior probability is determined by what we already believed beforehand, and want to focus on the contribution of the evidence, then we can say that the evidence S provides better support to G than to ~G if and only if P(S|G) > P(S|~G). In fact, the problem of evil claims, P(S|~G) is much higher than P(S|G), so the existence of suffering provides powerful evidence for ~G.
Bultmann, Rudolf. New Testament and Mythology and Other Basic Writings. Augsburg: Fortress Press, 1984. [The essay "New Testament and Mythology" was first published in 1941.]
Goodenough, Ursula. 1999. "The Holes in Gould's Semipermeable Membrane Between Science and Religion." American Scientist, May-June.
Gould, Stephen Jay. 1997. "Non-Overlapping Magisteria." Natural History 106: 16-22.
Robinson, John A. T. 1963. Honest to God. Philadelphia: Westminster Press.
Tillich, Paul. 1957. Dynamics of Faith. New York: Harper & Row.