David Berlinski - The Devil's Delusion

A single power assumption is at work throughout: Unlikely events require an explanation. The power assumption trails in its wake two additional assumptions. The first is that old standby: The universe is unlikely. And the second has long stood by: If God created the universe, He must be more unlikely than the universe He created. From the power assumption and its sidekicks, an infinite regress very quickly arises, one in which God requires an explanation, which in turn triggers the demand for yet another explanation, and so another God.

It follows that if God created the universe, there are Gods stacked up behind him, each one creating the God below. Either we must give up on him (formerly Him) or we need to get to the God who really gets things going, and since there seems by this argument to be infinitely many of them, each presumably more powerful, and certainly more intimidating, than His Subordinates, this is an inquiry guaranteed either to fail or to lead to a revival of an especially vigorous form of polytheism.

Imagine addressing prayers to our fathers which art in heaven, as in one of those ghastly children’s books in which Heather has three mothers and Jamal a dozen fathers.

Depressing, no?

The demand that explanations mention only events no more improbable than the events they explain is in any case intolerably abstemious. “How often have I said to you,” Sherlock Holmes observed to Watson, “that when you have eliminated the impossible, whatever remains, however improbable, must be the truth?”

But if we explain an event by an appeal to an improbable event, it does not follow that we are bound to keep clambering up the ladder of an infinite regress. When in The Perfect Storm Sebastian Junger described a freak storm off the coast of Nova Scotia, he was explaining shipwreck at sea by means of a very rare confluence of meteorological factors. Such explanations are common throughout the sciences and they are common in ordinary life. About such rare events, all that we can say is that sometimes they happen. We do not need to say more. What more could we say? It may well be that God is improbable and that is the end of it. When Christian believers give thanks for the miracle of Christ, they mean by a miracle a miracle.

Although Richard Dawkins has nothing but contempt for theology, often glorying in his impressive ignorance, with his argument he finds himself occupying an unexpected position of prominence amid “the peculiar debris of an abandoned and virtually forgotten science.”

In addition to suffering the infirmity of improbability, the God whose existence Dawkins is prepared to challenge seems curiously a diminished figure. He has gotten the job of creation done. His time thereafter has been spent imposing onerous sexual constraints upon the Jewish people and when absolutely required undertaking a miracle or two. For the moment, He seems to have vacated the universe with a smashing headache. On his previous appearances, He seemed very much like a lumbering robot. One might almost expect to hear the lingering echos of divine clanking. Above all, He is very much a contingent deity. If He is here today, He may be gone tomorrow. If His existence were guaranteed, the argument that Dawkins has advanced would fail before it started instead of starting before it failed.

And yet these are considerations long familiar in the history of theological thought. They form the heart and soul of Aquinas’s second cosmological argument, and if Aquinas gives them only a few words, that is because he requires only a few words to say what needs to be said. Any conception of a contingent deity, Aquinas argues, is doomed to fail, and it is doomed to fail precisely because whatever He might do to explain the existence of the universe, His existence would again require an explanation. “Therefore, not all beings are merely possible, but there must exist something the existence of which is necessary.”

The conclusion that a religious believer will take from Dawkins’s argument is either that God is improbable or that He is necessary.

What Dawkins has established serves chiefly as a reminder: Explanations come to an end, and because we are human, they must come to an end before they have satisfied every one of our emotional needs. But scientific atheists should at least be open to the possibility that scientific explanations by their very nature come to an end well before they have done all the work that an explanation can do. If they have not read Aquinas’s Summa Theologica, physicists have nonetheless heard its music. They have hoped to discover laws of some final physical theory so powerful that they will explain the property of matter in all its modes. “The most extreme hope for science,” Steven Weinberg has written, “is that we will be able to trace the explanation of all natural phenomena to final laws and historical accidents.”

This is the most extreme hope for science for those, like Frank Wilczek, inclined to say at some point that that’s just how things are. For others, intellectual comfort is less easily purchased. “We feel,” Wittgenstein wrote, “that even when all possible scientific questions have been answered, the problems of life remain completely untouched.” Those who do feel this way will see, following Aquinas, that the only inference calculated to overcome the way things are is one directed toward the way things must be.

Perhaps in the end this will prove to be a matter of mathematics. MIT physicist Max Tegmark has argued that this is so. The physicist Edward Witten and the mathematician Alain Connes have both written suggestively about the origins of creation in some inexplicably austere and remote mathematical structure, one so powerful that from it space and time themselves may be derived.

With these ambitious speculations on the table, it is worthwhile to recall that in locating the origins of creation in some fundamental abstract structures, mathematicians are assigning to them a degree of agency that until now they do not seem to have possessed.

There remains a final point. What a man rejects as distasteful must always be measured against what he is prepared eagerly to swallow. What Richard Dawkins is prepared to swallow is the Landscape and the Anthropic Principle. The Landscape does not, of course, answer the question what caused the Landscape to exist. How could it? And if nothing caused the Landscape, it does not answer the question why it should be there at all.

But having swallowed the Landscape with such inimitable gusto, Dawkins is surely obliged to explain just why he scruples at the Deity. After all, the theologian need only appeal to a single God lording over it all and a single universe—our own. Dawkins must appeal to infinitely many universes crammed into creation, with laws of nature wriggling indiscreetly and fundamental physical parameters changing as one travels from one corner of the cosmos to the next, the whole entire gargantuan structure scientifically unobservable and devoid of any connection to experience.

This is a point that Dawkins endeavors to meet, but with markedly insufficient success. “The key difference between the radically extravagant God hypothesis,” he writes, “and the apparently extravagant multiverse hypothesis, is one of statistical improbability.”

It is? I had no idea, the more so since Dawkins’s very next sentence would seem to undercut the sentence he has just written. “The multiverse, for all that it is extravagant, is simple,” because each of its constituent universes “is simple in its fundamental laws.”