Sunday, November 3, 2013

Divine action

by David Corner

It is important to theistic religion that it conceives of God as being active in various ways. God creates the world, and human beings; he is also thought to perform miracles. On the usual picture, everything God does in the world is something that God causes to occur. But God is not normally conceived as being part of the natural world himself. He is supernatural. This means any event in the world that expresses God’s activity is understood to be the effect of a supernatural cause. Let us call this view- the view that God acts by supernaturally causing things to happen in the natural world- supernaturalism.

I have trouble getting a clear grasp on this idea of supernatural causation. I think I have a handle on the notion of a cause generally, though maybe I am fooling myself here. But for example, I think I understand the following claims:
  • Sally caused her car to start by turning the key in the ignition.
  • Sally caused Betty to get a sunburn by leaving the tanning bed on for too long.
  • Sally caused the chandelier to fall by cutting the chain from which it was hanging.
  • Sally caused the vinegar to foam up by putting baking soda in it.
All of these claims have something in common. First, they are all instances of what is called “event causation,” by which one event, a cause, brings about a second event, which is the effect. So for example, the event (1) represented by the key’s turning in the ignition of Sally’s car causes the event (2) of the car’s starting. But also: Sally is a human being, and so she has a body, which is a physical thing. And these kinds of causes fall under familiar principles: There are mechanical causes, electromagnetic causes, chemical causes and so forth. Furthermore, in these cases, the causal relationship exists between physical things that have physical properties, like mass, wavelength, charge, chemical valence, and so on.

But God is usually thought of as being able to cause things to occur without having a body. Furthermore, God is conceived to be all-powerful. So if God exists as theistic religion supposes he does, he could start a car, give Betty a sunburn, make a chandelier fall, or make vinegar foam up. How can he do all these things without a body? Remember, God is not a physical thing, and so he has no physical properties. He has no mass, no charge, no wavelength, no chemical valence.

A supernatural cause has, as far as I can tell, nothing in common with any other sort of cause. Thus it seems to me to be an empty notion. Why call this “causation” when it bears no resemblance at all to any other sort of cause? Suppose I told you that I had a pet bird called an Oolumph. You are interested; what sort of bird is this? You have never heard of such a thing. Does it have wings, feathers? Does it lay eggs? I inform you that it has nothing in common with any other birds with which you are familiar. You would be within your rights to ask me why I call it a bird.

I suppose we could say that supernatural causation is just its own kind of cause. (This would be like suggesting that the Oolumph is just its own kind of bird.) A particularly inviting possibility might be to deny that supernatural causation is a form of event causation. But thinking of God’s activity in causal terms invites confusion. For instance, theistic philosophers suppose that God created human beings- i.e. that he caused them to exist. But then they often suppose that this account of the origin of human beings competes with the account given by evolutionary biology, so that only one of these accounts can be correct. They seem inclined to suppose so because they think the word “cause” is univocal (i.e. means the same thing) in these two sentences:
  • Human beings were created (i.e. were caused to exist) by God, and
  • Human beings came about (i.e. were caused to exist) by evolutionary forces.
It seems to me that taking these two claims to be in competition with one another has been responsible for a great deal of trouble. If I insist that my Oolumph is its own kind of bird, I should not enter it in any bird shows.

The problem of divine agency that emerges here is a familiar one to philosophers. We encounter it in what is known as the problem of mind-body interaction, which arises in regard to a theory of mind known as substance dualism. It appears as though mind and body interact in various ways; the physical event of hitting my thumb with a hammer, for example, causes a mental event to occur, namely the sensation of pain. The mental event of my willing my arm to go up causes the physical event of my arm’s going up- or so the substance dualist supposes. But the substance dualist denies that the mind is a physical thing. How is this sort of causal interaction possible if the mind is not physical? Supernaturalism seems to be a variety of substance dualism.

All this talk of supernatural causation presents a serious problem for the view that God acts in the world- and by extension for theism generally. What to do?

One solution is to abandon theism. Science tells us everything we need to know about what happens in the world; there is no room for God. Or perhaps there is a God, but God never does anything. Obviously neither of these is a very attractive solution to the theist.

I think there is another solution. I don’t have the space to fully explain it here, but perhaps I can give some general indication of the direction that I think theism ought to take. The mistake that supernaturalism makes is in supposing that God can only act by being the cause of events in the natural world. It’s tempting to suppose this is true of agency generally- that anything an agent does, she does by causing something to occur. This is false.

Some time ago, a philosopher by the name of Arthur Danto noticed that some of the things we do are not things we cause to occur. He referred to these as “basic actions.” This notion received quite a bit of discussion from other philosophers, such as Donald Davidson, Alvin Goldman, and Jennifer Hornsby. Davidson refers to these as “primitive actions,” and says on p. 49 of his book, Essays on Action and Events, that “event causality cannot…be used to explain the relation between an agent and a primitive action.” Suppose that I move my fingers, thereby flipping a switch, which causes a light to go on. We would rightly say that I caused the light to go on, and that I caused the switch to be flipped. But I do not cause my fingers to move. I simply move them. This is a basic action on my part.

This means that we can speak of an agent as performing an action without making any references to causes.

We have seen that serious problems are implied by the view that God supernaturally causes anything to happen. But if God exists, we can say that God acts in the world in a basic sort of way. There are things that God “simply does,” in the same way that I move my fingers. These are not things that God causes to occur. I suggest that theists abandon talk of supernatural causes and speak instead in terms of basic divine action.

As a postscript, I regret to report that Arthur Danto died just a few days ago, on October 25th, 2013, and I would like to provide a link to his obituary in the New York Times. Though he wrote on many topics, he was best known for his work in the philosophy of art. I had the pleasure of conversing with him on a couple of very memorable occasions.

David Corner
Senior Lecturer
Department of Philosophy
Sacramento State


  1. Thanks for the post professor Corner. A few things:
    (1) There are several examples in physics of efficient causation without material causation. One particular example would be the expansion of space. Space itself is expanding, not the material cosmos expanding into empty space. Some models of the beginning universe also seem to suggest an efficient but not material cause. And if the notion of abstract objects is coherent, then things like Beethoven's 5th are not physical objects and thus have no material cause, yet they began to exist and so we seem to be able to stipulate an efficient cause. Thus, it seems that God's being nonphysical poses no such problem for supernatural causation in terms of not resembling other forms of observed causation.
    (2) Another option open to the theist besides (i) atheism, (ii) denying supernatural causation, and (iii) basic action would be (iv) panentheism or process theology which has very unique metaphysical ideas about how God relates to the universe and seems to reject classical supernaturalism.

    1. Hi Miles. Thanks for your comment.

      Some questions in regard to (1): I'm sure the expansion of space has a role to play in a number of causal interactions; I suppose, for example, that this is what is responsible for the fact that the galaxies are all getting farther apart. Even if the expansion of space is not to be understood as an expansion of matter, wouldn't it still be physical?

      Also, if Beethoven's 5th is an abstract object, did it begin to exist? I mean, the abstract object itself rather than its instances in, say, the paper it was written on or the sounds that occur when it is performed. And in what sort of causal interactions do you see it playing a role? Did it cause Beethoven to write it down, or did Beethoven cause it to exist? And if Beethoven caused it to exist, how did he do that? By writing it down?

      As for (2), I don't see panentheism as an alternative to an account of divine actions as basic. The panentheist still has to give an account of what it is for God to act, and anyone who acts (human or divine) must engage in basic actions.

      Panentheism is actually quite friendly to the idea that God acts in a basic way, because it's so easy to understand the analogy to human action. To say that God acts in a basic way in the universe is to say that God has the same relation to the universe that I have to my body. This is quite easy to make out, at least for certain forms of panentheism, like Ramanuja's; he held that the universe is the body of God.


    2. Prof. Corner,
      Thanks for responding to my initial comments. Firstly, I'm unsure I understand your question about the expansion of space. It's probably also the case I don't fully understand the notion of the expansion of space. Admittedly, I'm going off of second-hand testimony with regards to this being a case of a phenomenon for which the cause is efficient but not material. You seem to be implying that the expansion of space itself is a physical cause of other phenomena but that wasn't my point, unless I've misunderstood you. I think there are also cases, if I'm not mistaken, from QM of this notion of efficient but not material causation being posited (perhaps in cases of quantum indeterminacy, or something?). I'm certainly not qualified in this area, so I'll concede that I was mostly blowing smoke. I don't intend to defend abstract objects, either, and would just as soon drop that approach in favor of the empirical, scientific ideas above. I think the point still remains that there are likely examples in our experience where we posit an efficient but not material cause for phenomena, and this shows that although perhaps difficult to understand, God's action as a nonphysical agent can't, in light of these, be considered to bear no resemblance whatsoever to any sorts of causes we are familiar with (unless you're not familiar with them).
      As far as my point about panentheism is concerned you're certainly right that basic action is consistent with this notion and I was mistaken to have considered them mutually exclusive. Admittedly, I was more interested in the claims of the first half of this post than the latter half and didn't take the time to appreciate the idea of basic action you develop.
      Although I have yet to read it, I found an article today by Daniel von Wachter titled "Do the Results of Divine Actions have Preceding Causes?" (2011) in which he apparently draws upon this idea of basic action as well. I wonder if you're familiar with it, and if so if you agree with what he has to say (or perhaps this isn't the place to delve into that).

    3. Miles, you say that "there are likely examples in our experience where we posit an efficient but not material cause for phenomena." I'm not sure I understand what you mean by 'material cause' here, though I am supposing you mean to say that there are phenomena that have no physical cause.

      It is true that there are (as I understand it) phenomena that have no physical cause- and certain quantum phenomena are among these. But then, do these phenomena also have efficient causes? If so, what are they?

      I just don't know what it would be for a natural phenomenon to have a non-physical cause. Are you suggesting quantum events are an example of such a thing?

      I am not familiar with Wachter's article, but I will search for it. Thanks for the tip!

    4. Miles, I've looked over Wachter's article but only briefly. I'll have to give it a closer reading. But I think he has it pretty much right.

  2. David, I think Miles' point is an interesting one. I have never been quite comfortable with the claim that causal interaction between fundamentally different substances isn't possible. When we ask, for example, how a mental substance can interact with a physical one, we are suggesting that there is something incoherent about it. But there is no logical contradiction in simply supposing that they do. To me the problem is just that this supposition isn't fruitful. Supernatural causation isn't testable, and therefore can't produce new knowledge.

    The idea of a basic action is interesting. In human agents, it strikes me as simply an admission of the limitations of he explanatory reach of the agency framework. In other words, to the extent that we conceive of a certain subset of motions that human bodies make as actions, we conceive of them as uncaused. But of course, these motions are caused, so if you want to understand their cause, you are going to need access to a different framework, e.g. a neurobiological one. I would be inclined to apply the same thinking to God's agency. In which case we haven't removed the problem of supernatural causation (if it is a problem) we have just quarantined it within a particular unproductive explanatory framework.

    1. Hi Randy.

      I tend to agree with you- I have no argument that interaction between substances of different attribute is not possible. It's just that I have no idea how they could interact; to use 18th century language, I can form no adequate conception of it. I think the claim, "God supernaturally caused x to happen," if it is meaningful at all, just reduces to the claim that God did x.

      What you say about basic actions is entirely correct. When I move my finger, this is a basic action on my part. I have no causal relationship to the movement of my finger. But as you point out, the movement of my finger (considered now as an event rather than as an action) does have a cause.

      What this implies is that something's having a natural cause doesn't rule out the possibility of attributing it to someone's agency. My moving of my finger (understood as an action) is, I would like to say, *realized* in a chain of events that can be given a natural explanation.

      So by the same token, we might say that God's agency in, say, the creation of human beings is realized in a series of events that can be explained by evolutionary biology. I argued for this in a paper that I presented at an American Academy of Religion meeting a while back.

      I view this as the most interesting implication of the notion of divine basic action. I'm very interested in the possibility that belief in divine agency can be made consistent with naturalism; hence the title of the book I am working on: "Naturalized Theology," which is a play on "Natural Theology." We'll have to talk about this some more.


    2. You have offered a fine, deep treatment of divine action. Your latest point—the idea that “belief in divine agency can be made consistent with naturalism”--is also very interesting, especially because so many people would say the belief is not consistent with naturalism. I wonder, though, whether you want to set your sights even higher than mere consistency and instead establish that belief in divine agency is required by the best explanation. All the observations made by all humans throughout history are consistent with the moon being made primarily of green cheese [and astronaut reports and subsequent geological analyses of moon material to the contrary are due to hallucinations by astronauts and geologists], but the best explanation of those observations is that the moon is not made primarily of green cheese. Establishing the consistency with belief in green cheese doesn’t add much to our understanding. Finding that the best explanation is that the moon is made of certain earth-like rocks does.

    3. Brad, this is an interesting point as it relates to the theory of agency. It seems to me that with respect to human agency, philosophers want to claim that intentional explanation constitutes a level of explanation that is not eliminable, i.e., that even though we can allow that all human motions have physical causes, there is a subset of such motions that have the quality of being actions, and we can not understand the nature of actions except by reference to intentionality. So if God is a parallel case, David's aspiration should be to show that there is a subset of natural motions the proper understanding of which requires reference to divine agency. Of course, if you are an Occasionalist, you believe that human agency is divine agency, so the subset of motions requiring human intentionality actually require divine intentionality. That's a yawner, though. What would really be interesting is to show that there is a distinct set physical motions that me must understand as Divine actions. Perhaps miracles are a candidate?

    4. Thanks Brad. As you note, many people do say that belief in divine agency is inconsistent with naturalism; in fact, to the best of my knowledge, I'm one of very few who wish to deny this. My primary concern so far has been to relieve theologians of the impression that they have to mount a war on science in order to maintain that God acts in the world. Conversely, the success of scientific explanations does not threaten belief in divine agency.

      Your challenge is well motivated, however: In what way might references to divine agency add to our understanding of the world?

      I understand Randy to be taking up this question. This is the point at which I hand the issue off to theology. If I am right, the theologian may acknowledge, for example, that evolutionary biology gives an adequate account of the development of human beings, where "adequate" is defined in scientific terms. In this acknowledgement, she parts ways with theists who feel they are committed to arguing for the scientific inadequacy of evolutionary biology, e.g. the Intelligent Design folks.

      The theologian is still free to insist that a proper understanding of certain physical motions requires us to appreciate their status as divine actions, and yes, miracles would certainly be a candidate. So in speaking of miracles, let's take the parting of the Red Sea as an example. This is a phenomenon that some have suggested might have a scientific explanation. And the naturalist-theologian need not deny this. But, she may argue, to suppose this explanation says everything that needs to be said about the parting of the Red Sea is to miss the significance of that event in salvation history.

      Of course the issue now is, why should anyone be concerned with salvation history? I don't know of any non-circular way that the theologian can argue for the importance of this.

    5. As a postscript to my last response, let me add that my reference to salvation history is a particular kind of theistic response- probably most recognizable to Christians. I think there are other ways of making the same point. But all of them will presume the value of theism.

      I don't see any way to appeal to the appearance of divine agency, e.g. in the creation of the universe, in the development of human beings, or in the occurrence of miracles in order to establish the existence of God in a way that has any chance of satisfying a religious skeptic.

      I am reminded of a remark Wittgenstein made, when he said that a miracle is like a gesture that God makes. Understanding a gesture, like a bow or a wave, requires a certain kind of context- a framework, one might say. The significance of a miracle, as an instance of divine agency, can similarly be appreciated only within a certain kind of religious framework. The occurrence of the miracle can't establish the legitimacy of the framework; on the contrary, the framework comes first and is what allows us to understand a set of physical motions as an expression of divine agency.

  3. That's very interesting Dave, thanks.