By S. Harris

What does it mean to explain something? To do it, you have to give some new information about it — how it works, how it came about, what it’s for, etc. To explain how a computer works, you might talk about, say, microchips or information processing — both explanations though different ones. To explain what a computer is, to someone who’d never seen one, you’d probably tell them what it does and what one looks like. These would also overlap with explaining where it came from (humans made them) and what it’s for.

Computers were made by people who more or less knew how they would work, and of course what they were for. If we’re explaining a natural phenomenon, or something done by a foreign people, we’ll have to find an explanation before we can give it. It will still have the same requirements: if we see something (say, that the sun radiates lots of energy), then we have to tell something new to explain what we see (that the sun is full of nuclear reactions).

What has been called the virtus dormitiva fallacy involves explaining something without really explaining anything. The name comes from the satiric explanation for why opium makes you sleep: because it possesses a dormitive principle, virtus dormitiva. So if you paraphrase both sides a bit, you get: it has the power to make you sleep because it has the power to make you sleep. Nothing is explained, yet if you do this with sophisticated enough words and add more layers of reasoning, it may sound like something has been.

Even philosophers who work with analysing concepts have been known to fall into this trap. What makes it more likely is also that sometimes people will not accept a real explanation.

Nicholas Maxwell (especially in The Human Mind in the Physical Universe) has said that the most important question in philosophy is how to reconcile the mechanistic, scientific view of the universe with the human view of the universe as containing meaning, value, and minds. Not that it’s difficult, mind you, but it seems it doesn’t come naturally to most people. Most of the non-explaining explanations I’ve seen have to do with this question. People are afraid to cross the border.

Take the classic question: where did the organised complexity in the universe come from, given that otherwise there is only cold physics? Ingenious explanations are provided by evolutionary and complexity theory, which have found ways in which simple physics (even mathematics) can organise themselves into complex phenomena without any pre-existing complexity guiding them. Anti-explanations are provided by those who deny this is possible and postulate that it must be a miracle or otherwise the work of an existing supreme being capable of creating and designing such complexity. So how did it happen? Miracle! Where did the god come from? Doesn’t need explaining!

(It should go without saying that the idea of a scientific explanation and the idea of a supernatural creator don’t have to conflict for a religious person. Yet, there are plenty of those who think as I described.)

So what is the meaning of life and how can there be any? What is morality? You can start to explain such things with a scientific world view as a basis. You can also say that meaning in life comes from god and morality means that you should do as god says. But how does god give meaning to anything? Why did a supreme being need to create this world with its meanings anyway? If you can answer these without going round in circles, you could probably answer the same questions without god too. And as for morality, “should” is just another word for “it’s morally right” — it’s a paraphrase, not an explanation.

Other such refusals to take an explanation might involve refusing an analysis of free will as either deterministic or indeterministic, even though “indeterministic” just means “not deterministic”, so there is no third option. When you analyse it, it just doesn’t sound like the same thing any more. So some philosophers have even tried to defy the logical contradiction and formulated “theories” in which it’s neither, and is actually contained as an unanalysed whole. No answers to anything there. In fact, such “explanations” are worse than none at all. Not only do they tell nothing, they also slam straight into a dead end. And worse still, they pretend they don’t.

The above example brought out an important thing: Once analysed, it just doesn’t feel like the same thing any more. That’s probably the core of most anti-explaining. It’s probably a psychological thing: we have a quick, unanalysed way of thinking about something, such as what counts as a free choice, and when we hear the explanation, that way of thinking is not activated and it doesn’t feel right. Such overly simple appeals to intuition should not be accepted. Remember, when we are explaining something, we cannot just say the same thing again in different words. We cannot just say there is a virtus dormitiva. This means that the explanation must be different from the thing that it explains. The fact that this makes people uncomfortable does not justify getting stuck making ever more elaborate explanations to match such “intuitions” so that their unstated requirement is that the thing being explained must be hidden somewhere as an unexplained whole.

An example of such an unexplained whole is the “homunculus” effectively postulated by some theories of the mind: you explain what goes on in the brain (or whatever) in some detail, but hidden inside your model of a mind, there is actually another mind. So you might explain vision in a way that in the end means that there is someone inside your head watching the pictures. A real explanation would explain the whole process as made up of parts none of which alone has the ability to do what the system does.

This is why I speak of anti-explanations: it’s when you want a non-explanation because you specifically won’t accept a real explanation.

One more example from philosophers, one containing some irony: modal realism, or concrete possible worlds. David Lewis was trying to answer the question of what we’re talking about when we talk about possibility, and his answer was that it’s a good explanation that every possibility happens in another world, a parallel reality just as real as this one but not accessible from here.

Most people do think this is crazy. Ironically, some of the counterarguments Lewis answers in his article are also refusals to accept an explanation. The objection is that this is not what it feels like we mean by possibility. I mean, whoever thought we were talking about other worlds rather than possibility in this world? If that were the only problem, I would be entirely behind Lewis on this one. Again, an analysis of possibility should be expected to reduce it to something else. There’s just the problem that this is an entirely useless analysis, and in a way it itself is a big refusal to analyse possibility through something else.

The theory of concrete possible worlds is an extreme example of reification, making an abstraction into a concrete thing. The biggest problem it has is simply complete lack of evidence. How does our speaking and thinking of something we call possibility suggest that there are parallel universes? There is absolutely no connection. It’s true in metaphysics we can postulate sort of objects to make our theory neater, but these are things like “redness in general”, never concrete things. They don’t “exist” in the same way.

But the real problem in terms of this theory as an explanation is how it explains precisely none of the concrete facts. It merely tries to create a structure in which our speech has a referent — and, in spite of being totally outlandish, does this in an extremely unimaginative way. When we are talking about possibility, we are talking about exactly the same kinds of things as we are talking about otherwise. They’re just somewhere else. But that “somewhere” else is another universe with no causal connection to ours. Thus, the only difference it would make if none of these other universes existed would be that our talk of possibility would be false according to this particular theory. How we actually come to know about different possibilities doesn’t have anything to do with seeing these other universes. The reasons we talk about possibilities have to do with this world. We still need a theory of how we talk about possibility and what we in practice mean by it, even if modal realism is true. All modal realism achieves, then, is not having to analyse how our talk about possibility really works, by claiming it works just like talk about more concrete things.

An explanation shouldn’t be completely foreign to the thing being explained; it must still explain the same original thing. But to determine whether this is so, what you need to do is analyse what in the original thing needs explaining — not just test the explanation through an unanalysed intuition.

Putting aside philosophers now, there is also a tendency towards anti-explanations called the fundamental attribution error: we tend to assume that other people, especially those we don’t know, do what they do because of their character, not the circumstances. Of course, it can be the best explanation, but the fundamental attribution error is an unrealistic bias towards such explanations.

Anti-explanations often overlap with the supernatural in the sense of that which cannot be humanly known and explained. Such supernatural explanations also have “and then a miracle occurs” embedded in them somewhere — a point where it just is like that and there is no further explanation. Read more about that here, or in Finnish and in relation to science here.

It has to be said that explanations must stop somewhere. The very last level is thought to be that of ultimate natural law. If Newton’s physics were completely accurate, that would presumably mean that gravity is something that can no longer be explained. Some objects just have a quality (mass) that makes them attract each other. Einstein’s theory of relativity showed that it’s more complicated than that, but it just found another layer of explanation underneath and stopped there. It is pretty commonly assumed that there is some ultimate level of natural law (“theory of everything”) explaining relativity and quantum mechanics and not explained by anything further. It’s a reasonable guess. At this level, then, we’d be stuck with something like a virtus dormitiva: it just is like that. But since that level is so ultimate we haven’t even found it and most questions are so complex it would take a lot of effort to get from there to the theory of everything, there’s a lot of explaining that can still be done. Indeed, usually we won’t go anywhere near that level. You can explain why someone did something without going down to the level of quarks.

Explanations have to stop somewhere, and fundamental natural laws are not explicable. It’s possible, though not that likely, that we will even run into some totally new basic natural law such as one allowing immaterial souls. But you can’t do that lightly, and if you do want to explain something, you’re going to have to go beyond the thing you are originally explaining. It might feel different. It’s supposed to — though if you can get rid of that feeling, you’re presumably expanding your understanding.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s