Causes and Reasons

Jeffery Hodges left a comment on our last post about free will (and I do apologize for approaching the subject so circumspectly, over a period of weeks) in which he asked if I was making a distinction between causes and reasons. This is an important question — and indeed I am.

To be clear about the difference: reasons are understood by us to be normative and logical abstracta. If I am heading for work, and the sky is grey and heavy, I’ll take my umbrella, for the reason that I am likely to get wet if I don’t. Causes, on the other hand, are exactly what we are referring to when we talk about determinism: my brain sends a signal down my arm that causes my hand to grasp my umbrella.

Part of the confusion that arises in this subject has to do with an ambiguity about what moves us from one step to the next: we don’t mind, it seems, being forced to choose a particular option as long as the choice is the only “reasonable” one to make. I’ve just severed an artery in my hand. Should I apply direct pressure and call 911, or go take a dip in the bay? The choice to seek help is “forced” upon any sane person, but that sort of “determinism” is OK, apparently.

What does seem repellent to many is the idea that we act not in response to to reasons, but “merely” to a cause-and-effect chain taking place in our heads. C.S. Lewis, in his book Miracles, seized upon this, and called it “The Cardinal Difficulty of Naturalism”. If our choices are not driven by reason (what Lewis called a process of “ground and consequent”), but rather by “mere” cause and effect, then what grounds do we have for trusting them? How can we have confidence in the processes of our own reasoning?

This seems like a good case on the face of it, but it isn’t, really (we took a closer look at it back in April of 2006). The principle weakness in the argument is its absolutism; what Lewis overlooks is that we don’t need to be Godlike, perfect deliberators to get by, nor do we need to have confidence in the perfection of our reasoning abilities in order to rely on them on a day-to-day basis. We know from psychological observation that our cognition is susceptible to all sorts of illusions and intuitive error; once we know the way it’s all wired up to work, it’s easily fooled. But it has evolved, over a very long time, to give good, practical results and a whole lot more; that it isn’t perfect should come as no surprise.

So, yes, our mentation, fine-tuned as it has been over the eons to effectively track and model the “reasons” that affect our well-being, can proceed according to neurological cause and effect with results that aren’t perfect, but are usually “pretty good” – good enough to get us this far, at least.

Related content from Sphere


  1. bob koepp says

    OK. I’ll voice a couple “concerns” that I hope you can alleviate with a bit of clarification.

    First, in distinguishing reasons from causes, we don’t want the distinction to admit of no overlap; i.e., we want reasons to play a causal role, but one that’s distinguishable from causal processes where reasons don’t operate.

    Second, being forced (i.e., compelled) by reasons is, indeed, normative. We ought to submit to compelling reasons (but some of us, obviously, do not…). This normative element is entirely absent from “mere” causal processes (whether these are deterministic or probabilistic).

    Posted May 5, 2008 at 12:48 pm | Permalink
  2. Malcolm says

    Hi Bob,

    Right – this is the sort of objection Lewis is getting at also. Indeed, causal processes are not normative in and of themselves. But, to return to the example above, you can easily imagine bridging this gap, say, by building a machine that, when I’m leaving the house and it’s raining outside, pops up an umbrella for me to take. And the position of the Darwinian materialist is that our brains are just that sort of machine, on a far grander scale.

    Posted May 5, 2008 at 12:55 pm | Permalink