I must admit that when the nice folks at Templeton asked me to write 1500 words in response to the title question, my initial response was to wonder how I would use the 1498 words that would remain after my two-word reply “We can’t!” Science is empirical. Theology isn’t. We increase scientific knowledge and discoveries by taking bits of the world and doing experiments on them. But we cannot do experiments in theology. (I trust it’s clear already that I’m no philosopher of science. I do know that the real story is more complicated. Some scientists don’t do experiments at all but just… think really hard about the universe. Whether that means that some sciences are not empirical or that some scientists are actually philosophers is a question for another day.)
So how do we increase theological knowledge and discoveries? I’m not sure. On many traditional forms of theism, we cannot make theological discoveries on our own at all, because theological knowledge must be revealed by God. Other traditional theists insist that before we can attain theological knowledge, we must first grow in virtue and personal holiness. Whether one accepts these claims or not, it is surely a mistake, a naïve and disastrous mistake, to treat theological inquiry as just another example of empirical inquiry. For the theologian, that way lies madness—and Dawkins.
So I doubt that we can make progress in theology by looking more carefully at objects in the world, or by designing better experiments. Well-meaning theists have long sought to confirm the existence of God by claiming that certain specific empirical observations can only be explained by attributing them to divine agency. Surely the marvelous complexity of the human eye, or the cell, or the biome can only be explained by appealing to a God who intervenes in the natural order, tinkering with natural events? Theologians rightly criticize this sort of project as “God-of-the-gaps” reasoning, and its track record has so far been one of unremitting failure.
And yet… theology may not be an experimental science, but it is (or ought to be) a form of rational inquiry. Perhaps there are assumptions, conceptual tools, and patterns of reasoning that are common to theology and the empirical sciences, precisely because they are presupposed by all forms of rational inquiry. I can think of several candidates—the laws of logical inference come immediately to mind—but I wish to focus on a pattern of reasoning found at the very heart of the empirical sciences, and argue that it has also played a major role in theology, even theology of a very traditional sort. I have in mind abduction, also called inference to the best explanation (IBE). In particular, I want to think about how we can distinguish IBE from the God-of-the-gaps reasoning I criticized above.
IBE is certainly plays a prominent role in our ordinary, everyday reasoning. Consider: you come home from work one day only to discover that the box of chocolates you had been saving is empty. You conclude that the best explanation for this lamentable state-of-affairs is that while you were out, your shiftless, chocolate-loving housemate ate the whole box. Of course, other explanations are possible. Your landlord could have used his master key to enter your apartment and eat your chocolates. Or a particularly crafty thief could have broken in and eaten them without leaving any other trace. But all in all, the best explanation is that your housemate is the culprit. Similarly, IBE is also a fundamental feature of scientific reasoning. Every datum or observation can be explained by a vast array of logically possible hypotheses. Among the plurality of possible hypotheses, scientists prefer the single hypothesis that best explains the relevant data. (For more examples, and for a guide to some of the tricky philosophical problems posed by IBE, see the entry on Abduction in the Stanford Encyclopedia of Philosophy.)
Many theologians repudiate any role for IBE in theology, because they assume that IBE can only result in bad, God-of-the-gaps reasoning. But whatever the problems with the latter, theologians who sympathize with traditional orthodoxy should not reject IBE as such. Indeed, there are vital theistic claims and practices that depend on IBE. For example, on a traditional understanding of the miraculous, when we call an extraordinary event a “miracle,” we mean to say that God has brought about that event by means of a special divine action, instead of by means of ordinary causes alone. When we judge that some event is a miracle, we engage in IBE. Suppose you ask God to heal your terminal illness, and the next day you wake up cured. You infer that the best explanation for your sudden recovery is that God has healed you. Other explanations are possible (sometimes diseases go into spontaneous remission as a result of natural causes) but on the balance of the evidence, you attribute your recovery to an act of God.
My point is not that your attribution is correct. My point is that theologians who repudiate IBE altogether must either deny that miracles occur or else give an alternative account of how we attribute them, one that does not rely on IBE. The point generalizes to all religious experiences, insofar as they are regarded as extraordinary experiences in which God communicates with someone uniquely and immediately. Someone who judges that he is in the presence of God, instead of, say, in the grips of a mental illness, is making an inference to the best explanation of his unusual experience. Theists cannot give up on IBE as a valid form of theological reasoning without also giving up on the ability to call some events miracles or religious experiences—a high price to pay. More controversially, I would also argue that some of the most central Christian claims—for example, that Jesus is divine, or that God is Trinitarian—were formulated as a result of IBE. I accept that I cannot really defend that argument now, however.
How, then, can we distinguish legitimate theological uses of IBE from illegitimate appeals to the God-of-the-gaps? I confess that I am not entirely sure about that either, but the question deserves further thought. One possibility is that “IBE” and “God-of-the-gaps” are not two different forms of reasoning but different labels—one positive, the other pejorative—for the same form of reasoning. We use “IBE” to describe those explanations that, as far as we know, remain sound, and “God-of-the-gaps” to describe those explanations that we now know or suspect to be false. Another possibility is that we can distinguish IBE from God-of-the-gaps at the level of the explanandum, by distinguishing the phenomena that they each purport to explain. Recalling the examples above, perhaps it is legitimate to appeal to special acts of God in order to explain extraordinary, unrepeatable events, but not ordinary, observable features of the natural world, like the complexity of the eye.
To return to the central question, I still don’t think that we can increase theological knowledge and discoveries “in the same way” that we increase scientific knowledge and discoveries. In my view, theology is more like metaphysics than physics. (I pause to note that this claim does not commit me to the view that it is very much like metaphysics…) Properly theological claims are rarely open to empirical falsification. But if it is a mistake to treat theology and natural science as too similar, it is also a mistake to treat them as utterly different. They are both forms of rational inquiry, and so they share certain features of rationality as such, features that are common to any form of rational inquiry. Inference to the best explanation is one such feature.
One way to make progress—or, at least, avoid regress—in theology is to continue to think carefully about how theological claims differ from, and are similar to, scientific claims. We should not conflate the two, but nor should we consign them to utterly separate spheres. In particular, we should think more carefully about what distinguishes IBE from God-of-the-gaps reasoning, so that we will have a better idea of how to employ the former while avoiding the latter. Some theists would deny that theological claims can ever be explanatory, even in principle, and they would resist all talk of “theological data” or “theological hypotheses.” But of course theology offers explanations, at least sometimes, and we need not subsume theology into science to admit that it can have its own data and its own hypotheses. We should worry about appeals to a “God of the gaps” because the track-record for such appeals is so poor. But we ought not take the further step of insisting that divine action can never be the best explanation of anything ever. That step not only cuts theology off from science, but also from our ordinary practices of reasoning. That step would make it seem as though theology is not really a form of rational inquiry at all.
- When, if at all, can we appeal to divine action as the “best explanation” of an event?
- What, if anything, distinguishes “God of the gaps” reasoning from inferences to the best explanation?
- In what sense, if at all, are appeals to the existence or activity of God “explanations”?
In my original piece, I first cast doubt on the suggestion that we can increase theological knowledge “in the same way” we increase scientific knowledge. I emphasized the fact that science is an empirical inquiry, but theology is not. Yet I also argued that both share important patterns of argument and modes of reasoning, and I identified inference to the best explanation (IBE) as one such pattern or mode. I also speculated about what, if anything, distinguishes “God of the gaps” reasoning from inferences to the best explanation.
The piece provoked a lively and high-quality discussion. One or two commenters were happy to bite the god-of-the-gaps bullet. They insisted that if some form of theism is true, then there must be gaps in our ability to give a scientific account of specific observable phenomena—consciousness was a popular example. To the extent that science is unable to give a purely physicalist or naturalist account of consciousness, theology gains; to the extent that the converse is true, theology loses.
I remain deeply uncomfortable with this line of argument. As I indicated in the original piece, my discomfort arises in part from induction on the record of similar moves made in the past: in the long term, when theists bet on the failures of science, they tend to lose. I am also not persuaded that consciousness is in principle unexplainable by science, as opposed to currently unexplained. Nor am I persuaded that theology (or at least, Christian theology—about which more below) really requires that consciousness arise from something immaterial or otherwise scientifically inexplicable.
Several commenters argued on various grounds, that theological inquiry is actually very similar to empirical inquiry. At the most fundamental level, they suggested, theology and science are both ways of asking the question “what is going on here?” They differ primarily with respect to the phenomena that they each regard as falling under the scope of the question.
To some extent, I agree with this line of argument. It is quite similar to my own proposal that IBE is a pattern of reasoning basic to both theology and science. I do insist, however, that the methods of theology and science are very different, because the latter relies on empirical investigation whereas the former does not. I agree that theology and science both lie on a continuum. As the questions asked by science get more-and-more general, they eventually resolve into properly theological (or metaphysical) questions like “why does anything exist at all?”
Whether expressly or implicitly, many commenters noted the Western and Christian assumptions that underlie my own piece, and they suggested that better answers to the title question might be found in Whitehead’s process theism or in various non-Western religions.
My own response to such suggestions is: maybe; maybe not; but it is not for me to say. I don’t intend that response to be flippant at all, and in fact, I think it is actually the most respectful thing I can say. I am shaped by my Western and Christian assumptions, and so I am interested primarily in bringing those assumptions to bear on the questions I was asked to address. Others would no doubt proceed differently. I don’t know enough about any non-Western religions to have informed views about how to bring their assumptions to bear. (I know a little more about process theism, actually, but that’s a topic for another day…)
One of the most interesting suggestions raised in the discussion was based on a point attributed to the philosopher David Lewis: that philosophy progresses by getting clearer on the conceptual costs of various positions. (As the commenter put it, “if we accept P, the cost is that we must also accept Q, which is unpalatable for reason R, etc.) Perhaps theology progresses in the same way?
I am inclined to think that it does, or at least that the suggestion captures one way that theology progresses. I also note—happily, from the point of view of the title question—that there is a sense in which science progresses in a similar way. When we test a scientific hypothesis, we try to ascertain the costs of holding it in comparison with rival hypotheses. The difference, as ever, lies in the kinds of tests we perform and the kinds of data we examine. But the idea that science and theology are both ways of asking the question “what is going on here?” and then trying to assess the costs of rival answers is a fruitful one.
New Big Questions
1. Is human reason universal, or does the fact that different cultures have different understandings of what is rational mean that reason is culturally constructed?
2. What is self-deception and why, if at all, is it bad?
3. What does it mean to “love the truth”?