Occam’s razor can leave you bloody

This will be a fairly short post — more of an excuse to rant than anything else. But I’m seriously tired of the common misuse of Occam’s razor.

Occam’s razor merely states that the simplest explanation is the best one. It’s roughly based on a law of logic that maintains that it is fallacious to posit more than what fully accounts for something as part of its explanation. This works extremely well for items of discernible composition: four quarters make a whole, and any more than that is not part of the whole, but something in addition to it.

What I object to is its use as an aesthetic masquerading as a principle of dispassionate logic. Too often people snip things away after unconsciously making assumptions about the composition of the thing being explained. Occam’s razor can easily cut out too much, and without analyzing those assumptions, it will never be noticed.

For instance, it’s often simply accepted that immediate causes and composition sufficiently explain. But that’s an aesthetic, not a precept of rational analysis: sometimes we would like to know more than the structural composition of a thing to understand it. Describing the raw materials is only enough when all you’re looking for is raw materials. But this is why Occam’s razor is useless for grander subjects: it frontloads the endpoint in question, styling the haircut (so to speak) after a preconceived, predetermined pattern. There are assumptions that go into what you think is necessary to explain things, and Occam’s razor is really handy for excising everything other than what you expected to find.

People wielding Occam’s razor to trim away everything but what they find personally interesting is not the same as saying that everything they trim is spurious, bogus, or otherwise worthless. When I walk into my back yard, see a shed that I didn’t know about, and ask my wife, “Where did this shed come from?” and she answers, “I got it from Big Lots,” I am fully enlightened only insofar that I wanted to know only the source of the raw materials. But it doesn’t answer, “Why is it here? Who assembled it? And where did the money used to buy it come from?”

Similarly, theologians and philosophers are always about answering more things than science answers. So when a materialistic scientists announces, “I’ve explained the natural causes of the universe! No need for the God hypothesis anymore!” we can safely say that they have taken Occam’s razor to everything but what most interested them.

And that’s ok–they can do a fine, parsimonious job of explaining that stuff. Just don’t let them try to tell you that there are no other questions worth asking.

Tagged with:
Recent Posts:
  • I’m confident that you know that technically Occam’s Razor is something like: “entities should not be multiplied beyond necessity.” I have often pointed out that “necessity” is a key word. It is when two theories have (at least) equal explanatory power that the simpler one should be preferred. (Hmmm… what if a theory explained 95% of the data, but with only 5% of the complexity of the alternative explanation? I need to think more about this trade-off.) Simple explanations are more beautiful than complex ones, but sometimes complex phenomena will require ugly, complex explanations.

    That said, I believe that Occam’s Razor cuts off the God hypothesis, and would like to hear more from you about why this is wrong. I definitely did not understand your critique of materialistic scientists.

    By the way, if Occam’s Razor is true, then it should be so for explanations of all phenomena – large or small.

    • Yes, Jared (good to hear from you). I am aware of that definition. As you anticipated, it really comes down to defining “necessity”. My critique is that what is necessary is too often presumed from the outset. If you are looking only for materialistic explanations of causes, you must excise everything else that is not causal, whether material or immaterial, in your solution. And Occam’s razor is good for that.

      But to your point about “the God hypothesis,” I’m not sure we’re talking about the same thing. I can imagine that what gets cut off is the old God-of-the-gaps type of cosmological argument, in which divine intervention is necessary to explain the composition of the material universe. As Laplace said, “I had no need of that hypothesis.” Naturalistic explanations routinely win the day for that sort of question. (Of course, it’s been sharply debated as to whether all the spontaneously arising laws of the universe and other intricacies of naturalism are really all that more simple than a single entity behind it all.) Anyway, if the two sides are the scientific explanation of material components/mechanisms vs. the theistic explanation of material components/mechanisms plus God, Occam’s razor does side with the scientific explanation.

      But what I’m saying is that committed materialists act as though this were the only debate worth having. We’re having deeper questions than that, always pushing things back further. Yes, that the universe is composed of matter and operates, by all we can see right now, under the laws of physics that seem to be somehow woven into the fabric of the universe is not in dispute (at least, by me). But the question of meaning is still worth pondering. My “God hypothesis” is that meaning and objective ethics are not illusory, emergent, or incidental but stem from something that grounds our universe; I fail to see how Occam’s razor I able to identify either position as having unnecessary “entities”. A self-bouncing pogostick has one fewer entity required than a pogostick with a kid on it, but is it really any more elegant or less complex?

      If I were looking simply for an explanation of our sense of meaning or why we feel like we need to look for purpose in the universe, I’d be satisified with, “We evolved that way.” That’s completely cool. But because we evolved that way, I can ask, “But why did we evolve that way?” Occam’s razor applied to the question of our material composition cannot legitimately lop off the inquiry of philosophers and theologians into that question of meaning that wise scientists have agreed to recuse themselves from.

      I agree that if Occam’s razor is true, then it’s valid for all phenomena. It’s just that our ability to quantify and analyze what’s simpler and what’s more complex diminishes as the complexity of our subject increases.

      • You said: “My “God hypothesis” is that meaning and objective ethics are not illusory, emergent, or incidental but stem from something that grounds our universe; I fail to see how Occam’s razor I able to identify either position as having unnecessary “entities”. A self-bouncing pogostick has one fewer entity required than a pogostick with a kid on it, but is it really any more elegant or less complex?”

        Apart from utilizing the same underlying laws of physics, an explanation for the motion of a self-bouncing pogostick will be different than that for a regular pogostick with a kid on it. Those are two different phenomena. Occam’s Razor could only help you choose between two competing explanations of the same phenomena. So, if a child is bouncing on a pogo stick, and someone attempts to explain the motion of the stick without reference to the child adding energy to the system, then one must ask whether the standard explanation or this new one explains the most data with the fewest entities. Like anything else, you really do have to set up the problem correctly in order to get a valid outcome.

        Your “God Hypothesis” seems like it should be testable on empirical grounds. What would a universe with morality woven into its very fabric look like (my words, not yours)? Similar or different to our own, and how? If the question of interest is something like, “Which type of universe is our own, impersonal and amoral or with a foundation that also grounds meaning and ethics?” then both sides should use data to support a theory, and then Occam’s Razor can be applied.

        I am really very curious… what evidence is there to support your view that meaning and ethics are not emergent phenomena?

        By the way, it is quite true that simplicity and complexity are notoriously difficult to define in this context. The concept of Kolmogorov complexity can definitely help here, but it only indirectly applies whenever we’re talking about something more than a data stream. Still, I find it very helpful.

        • Apart from utilizing the same underlying laws of physics, an explanation for the motion of a self-bouncing pogostick will be different than that for a regular pogostick with a kid on it. Those are two different phenomena. Occam’s Razor could only help you choose between two competing explanations of the same phenomena.

          You’re making my point. It’s no good saying that Occam’s razor can be used to distinguish a godless universe from a theistic one when all the discriminating factors are unavailable to us. Seemingly with every discovery, psychics is constantly showing how much more mysterious and complex our universe is–so how can we really hold up both and decide which one requires fewer multiplied entities?

          Your “God Hypothesis” seems like it should be testable on empirical grounds.

          Well, there’s a reason I put it in quotes. It’s not particularly testable empirically for the reasons I just suggested: on the surface it’s indistinguishable from a purely materialistic universe. It’s in speculating about what’s behind or not behind the universe that we’re both offering conjectures. Because 1) I simply don’t see any grounds on which to conclude that someone’s presupposed “morality is incidental” trumps someone else’s presupposed “morality is absolute” and 2) I find no grounds to conjecture that (e.g.) torturing children for fun could ever have been proper in any imaginable cultural context, I find an absolute basis for morality to be more plausible.

          The closest thing I have to “data” making my understanding more plausible to me personally is our experience with intentional agents vs. unintentional chance. Under what you propose, it might be folly to assume that a building discovered on an exoplanet ended up there by any but the most accidental of natural processes, since intelligent life has not been discovered or detected yet. Like God, it is possible to explain away our conjecture about the existence of aliens as projection and overapplied relics of an animal pareidolia instinct that has allowed us to become a dominant species. Rightly or wrongly, our hypothetical experience with a building on an exoplanet could prompt us to extrapolate the idea of something that before now has been completely conjectural as a cause for that building. Now, in the same sense that the building as constructed by alien agents would be fully “natural” (as a product of evolution) as well as agentive, the existence of morality, though mechanically explicable by sociological evolution, may not be exhaustively explained by immediate causes. As with the conjectured aliens, there may be something more complex and fascinating going on. It is on this that I think that Occam’s razor is more often used as a dismissive rather than a particularly robust analytical tool.

          But to be clear, I’m not pretending there’s any profound empirical basis for this; I’m not the one trying to prove anything. 😉

          • You’re making my point. It’s no good saying that Occam’s razor can be used to distinguish a godless universe from a theistic one when all the discriminating factors are unavailable to us. Seemingly with every discovery, psychics is constantly showing how much more mysterious and complex our universe is–so how can we really hold up both and decide which one requires fewer multiplied entities?

            Maybe our disagreement is an epistemic one, but I hope not. What I mean is, if all you are saying is that it is illegitimate to exclude the possibility of God based upon an Occamian argument, then I agree. Excluding the possibility is the same as saying we can know the opposite with certainty, which is not possible. The Razor does not deal in certainty, at least when it comes to the theories in question, but rather only in probabilities. So, while in my view the razor shaves off the God hypothesis, what I mean when I say this is that it shows it to be highly unlikely vis-a-vis the alternative(s). It could still ultimately be true, and could be shown to be so based upon new evidence. Then the razor would then shave differently. Faeries or phlogiston (or whatever) are still possible, even though the Razor has cut those theories off from their competitors. Probabilistic knowledge is still knowledge, however, and that means that the Razor is useful when used properly. I’m sure glad that most adults I know no longer believe in the evil eye.

            If you agree with the above, then how exactly do we legitimately exclude theories of origins from such analysis? When can the Razor be used if not in relation to the God hypothesis? It seems to me to be a poster-child for it (I’ll have to argue this point later). We can’t take this step just because new data may come in… that is always the case. Our universe is indeed complex, and we often find it mysterious (aside: this is more accurate than saying that the universe is mysterious, which is a seemingly innocuous category error built into our language that can end up causing a lot of confusion in these types of discussions – the universe is what it is regardless of how we feel about it). New discoveries happen all the time, and these must be integrated into existing knowledge. Sometimes this is easy, but often it requires adjustments to our old “true” theories, or even new theories. The quest for answers to the questions that interest us need not stop just because new and sometimes surprising data continue to come in. We just need to accept the fact that all of our answers are provisional.

            I wanted to get the above two paragraphs out of the way, because I don’t think that they are your real issue. I think you mention them because your thinking on the issue is muddled, and these get brought up because they are real limitations to rational processes, like the use of Occam’s Razor.* They may be annoying, but not your real problem with the Razor. I don’t want to put words in your mouth, so please correct me if I’m wrong. I’m projecting based partially on my own experience but also on our interactions in the past. I think your real problem lies in the seemingly very subjective way in which problems can be set up so that the Razor gives the desired outcome. I think you basically said as much in your post and comments. This is a very legitimate concern, but those other issues appear to me to be just distractions.

            I want to write a lot more, and respond to your thought experiment and the rest of your latest comment. For now, I am out of time. So, why don’t I leave it here and let you tell me if I am generally on target or way off base so far?

            I appreciate the time you are taking with this. I’ve wanted to have a conversation like this with a thoughtful theist for a while now.

            * Technically, you only mention the problem of fresh and surprising data, but it seemed like this rested on an underlying problem with probabilistic knowledge and provisional answers.