Occam's Razor

Every now and again, some atheist will claim that Christianity is falsified by Occam’s Razor. Occam’s Razor is the principle of parsimony, which states that entities should not be multiplied needlessly. Basically, the Razor claims that the simplest explanation is the best. The argument forwarded by atheists is generally along the lines either that (i) God is unnecessary to explain the world as we know it, and therefore is unlikely to exist; or, more strongly, that (ii) since God is infinitely complex, the Christian explanation of reality is thus infinitely more complex than a non-theistic one, and so should be rejected by default.

It intrigues me that atheists use this as a foundation for “disproving” Christianity. Several obvious problems suggest themselves:

Question-begging

Firstly, how does (i) not beg the question against the Christian? If, in fact, the Christian is correct in asserting that God is not just necessary to explain reality, but is a necessary precondition for reality, then (i) is obviously false and doesn’t constitute an argument at all. Since the Christian has plenty of good arguments of his own which seek to prove his position, these should be evaluated on their own merits rather than dismissed on the dubious basis of parsimony.

Less obviously, (ii) also begs the question. Even if the Christian explanation is infinitely more complex by merit of entertaining an infinitely complex being, perhaps it is the case that, in this particular instance, such a being is a requirement of any rigorous and adequate explanation of reality. The atheist needs to make an argument which shows this is not the case, rather than merely asserting it.

Furthermore, what does the atheist mean by “infinitely complex being”, in reference to God? The term “infinite” is used very freely with relation to God, but is generally a qualitative term rather than a quantitative one. That is, when we say that God is “infinite”, we tend to be referring to some superlative characteristic of his, rather than to any actual number of things which inhere in him. So the atheist needs to clarify and argue for his view that God is infinitely complex.

On top of this, even if that argument is successful, he has still not shown that an infinitely complex God entails an infinitely complex explanation. In what sense is the quantitative infinity of God being imputed to the Christian’s explanation of reality? Again, clarification and argument, rather than mere assertion, are required to prove the point.

Complexity is better than simplicity

Secondly, and along similar lines to the question-begging problem, it is self-evidently the case that we can have such a thing as an explanation which is too simple, but not necessarily an explanation which is too complex. Imagine, for example, a detective trying to find an explanation for the death of a man who died from blunt trauma in a factory. It’s obvious to us that an explanation which includes a murderer is more complex than an explanation which doesn’t. According to Occam’s Razor, the detective should favor any explanation which does not needlessly multiply entities. If the death can be explained by an unfortunate mechanical accident, then there isn’t any reason to postulate a murderer. A murderer becomes a needless entity, and so the detective assumes that it was indeed an accident. That’s fair.

However, two obvious things need to be noted: firstly, an explanation which fails to include a necessary entity is too simple, and therefore is necessarily false. Imagine the dead man was 90 years old and had a heart condition. Ordinarily, natural causes would be the simplest and most likely cause of death. But there is evidence of blunt trauma; so if the detective posits a natural heart attack as the explanation for man’s death, his explanation is obviously too simple—and thus must be wrong. A blunt object is a necessary entity in the explanation.

Secondly, and on the other hand, a murderer could have killed the man in such a way as to make the death appear accidental. So the fact that the explanation without a murderer is more simple does not guarantee its truth; and the fact that the explanation with a murderer is more complex does not guarantee its falsehood. In fact, we can imagine a fantastic and highly unlikely explanation for the man’s death, involving any number of entities that the detective would never think of, which was nonetheless true.

So an over-simple theory must be wrong, but an “over”-complex theory might be right. There are plenty of good arguments to show that a non-theistic explanation of reality is over-simple in such a way that it must be false. I hope to discuss more of these in the Philosophy section of Thinking Matters Talk as time goes on.

Occam’s Razor has no grounds in a non-theistic worldview

The last and most convincingly troublesome problem for the atheist is that Occam’s Razor itself, on which his objection is based, really has no grounds whatsoever in a non-theistic worldview. The atheist wants to say that we should not multiply entities needlessly. A Christian may well agree with him, because he knows from revelation (both special and general) that God typically does not act in a needlessly complicated way. He has designed the universe to act consistently, and in a way which is fairly straightforward, even in its complexity. He has also designed our senses and intellects in such a way that we can apprehend the way the world works, and discover things about it. Most importantly, he has built into us certain expectations about the world, such that our intuitions generally match up to reality. Thus we have grounds for affirming Occam’s Razor.

But an atheist has no such grounds. In a non-rational universe, whether mechanistic or probabilistic, what possible reason could he have for asserting that simpler explanations are better? Why should they be? As a rule of thumb, at least fifty percent of the time we should expect the more complex explanations to true. There isn’t any physical law of parsimony such that the universe must operate in such a way that simpler explanations are better, is there? So on what basis does the atheist assert Occam’s Razor at all?

He could say that, historically, the simpler explanations have been true. And maybe this is so. But then why does he think that this will continue to be the case? After all, we know very little of the universe, and we haven’t been around very long in the grand scheme of things. Perhaps our history is an aberration, and in fact it is a general rule that the likelihood of an explanation being true tends to rise with its complexity. How can he know this isn’t the case?

In truth, he affirms Occam’s Razor because his God-given intuitions suggest very strongly to him that it’s true. Unfortunately, because his intuitions are indeed God-given, he is most certainly misapplying them in using them as a basis for objecting to God’s existence.

1 reply
  1. Stuart
    Stuart says:

    Occam’s Razor, which is against needlessly multiplying explanatory entities should not be used as an argument against God’s existence. Richard Dawkin’s has recently made this mistake in his notorious book _The God Delusion_. He thinks that positing a divine designer for the explanation of the universe makes no explanatory advance, because the concept of God as an omniscient and omnipotent being is just as if not more complex than the universe itself.

    It is true that simpler explanations should be preferred, but one of the roles of the philosopher of science is to determine the role _simplicity_ should play in explaining events. It takes for instance great skill to weigh if simplicity should be abandoned for things like _explanatory power_, or _explanatory scope_, or degree of _ad hocness_, etc. In the end Dawkin’s confuses complexity and improbability with plausibility.

    But more fundamentally, the _simple_ fact is God is not complex. As a singular immaterial mind God (on the Christian concept at least) is startlingly simple. Alvin Plantinga, Christian philosopher at the University of Notre Dame makes this point powerfully in an article for the magazine _Christianity Today_ (March/April 2007, Vol. 13, No. 2, Page 21)

    >According to much classical theology (Thomas Aquinas, for example) God is simple, and simple in a very strong sense, so that in him there is no distinction of thing and property, actuality and potentiality, essence and existence, and the like. Some of the discussions of divine simplicity get pretty complicated, not to say arcane.3 (It isn’t only Catholic theology that declares God simple; according to the Belgic Confession, a splendid expression of Reformed Christianity, God is “a single and simple spiritual being.”) So first, according to classical theology, God is simple, not complex.4 More remarkable, perhaps, is that according to Dawkins’ own definition of complexity, God is not complex. According to his definition (set out in _The Blind Watchmaker_), something is complex if it has parts that are “arranged in a way that is unlikely to have arisen by chance alone.” But of course God is a spirit, not a material object at all, and hence has no parts.5 A fortiori (as philosophers like to say) God doesn’t have parts arranged in ways unlikely to have arisen by chance. Therefore, given the definition of complexity Dawkins himself proposes, God is not complex.

    *Footnotes:*

    3. See my _Does God Have a Nature?_ Aquinas Lecture 44 (Marquette Univ. Press, 1980).
    4. The distinguished Oxford philosopher (Dawkins calls him a theologian) Richard Swinburne has proposed some sophisticated arguments for the claim that God is simple. Dawkins mentions Swinburne’s argument, but doesn’t deign to come to grips with it; instead he resorts to ridicule (pp. 110-111).
    5. What about the Trinity? Just how we are to think of the Trinity is of course not wholly clear; it is clear, however, that it is false that in addition to each of the three persons of the Trinity, there is also another being of which each of those persons is a part.

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *