Saturday, April 5, 2014

Mailbag: An Argument against the Kalam?

Welcome to a special edition of the Possible Worlds mailbag! :)

Robby asks:

“Hey, I saw this today on the reasonablefaith forums and I thought I'd get your opinion.  It's supposed to be an argument against the KCA but I'm afraid I don't even understand what the problem is supposed to be.  Do you understand it?  Here it is:

Let C = the set of all things that have a cause.
Let G be an uncaused that is not a member of C.
Let A be some action take [sic] by A [sic] and which stands in causal relation to C.

1. C cannot have been caused by any member of C (even if everything else in C was caused by just one other thing in C, that one other thing could not be the cause of itself).
2. So, A cannot be a member of C.
3. So, A cannot have had a cause.
4. So, G cannot be the cause of A.
5. So, G cannot be the cause of C.”

Randy: Hey Robby, thanks for the question! So, it’s not clear what it means to say that A stands for some action and also an agent. That ambiguity may contribute to some problems. Charitably, the first sentence might be re-interpreted to mean “Let A be some action taken by G and which A stands in causal relation to C.” This disambiguates the definition some. However, the problem is that there’s a bit of ambiguity in what it means to be a cause. I’ll try to explain the argument, and then try to explain where it has gone wrong.

First, the argument appeals to set theory, in that the entire set of all caused things cannot be caused by a member of C. This is because in order to be a member of C, that thing (or member) must be itself caused. If some member of C were to cause all of C, it would have to be the cause of itself, which most philosophers take to be absurd/incoherent. So it follows that the action A, which stands in causal relation to C, is not a member of C.

Since C encompasses all caused things, then whatever is not a member of C is uncaused. That would include “G,” where G stands for God, and A, according to the argument’s presuppositions and the rules of entailment. From this it follows that G is not the cause of A (after all, how could something both have and not have a cause?). And since, according to the argument’s presuppositions, A stands in causal relation to C (presumably in the “causes,” rather than “caused by,” direction), and G did not cause A, then G did not cause C.[1]

I think an easy way out is to deny that A really is standing in causal relations to C, if we’re taking A as some distinct thing. It implies, in the presupposition, that God causes A, and A causes C (which is, presumably, the universe [or at least the set of all caused things]). But I’m getting ahead of myself. I’d like to argue that the above argument has conflated multiple types of causation. It implies that God is supposed to have caused each and every caused event (or agent, or thing, or whatever) involved in C. But why think such a thing? I, and many (probably most) Christian theists do not claim that God is the cause of everything that has a cause. What about agent causation, where those agents endowed with libertarian free will bring about some action? Now, I think it’s obvious the arguer has in mind some conception of causation that he or she wants to hold univocally throughout the argument, but I’m not clear on what it is supposed to be. And that is an extremely important point.

There are four Aristotelian categories or types of causes: material, formal, efficient, and final. Formal and final causes are plausibly out in this discussion (since “final” refers to telos, or purpose, and “formal” refers more or less to “the form” of something). So that leaves material and efficient.

To which is the argument referring? Material causation refers to “the material out of which the statue [this particular example] was made.”[2] So what is the efficient cause? The efficient cause is the “primary source of the change or rest . . . the artisan . . . the man who gives advice, the father of the child.”[3] The kalam cosmological argument refers to God as the efficient cause. William Lane Craig has noted, “An efficient cause is something that produces its effect in being; a material cause is the stuff out of which something is made.”[4] If God is the efficient cause on the kalam, then what is the material cause? As it so happens, on the kalam, the universe lacks a material cause. So, wholly apart from the fact that every non-theological causal determinist will deny that God causes or is involved in an unbroken causal chain with every caused event as its efficient cause[5], the causation involved in the kalam is solely as efficient causation of the universe.

Taken this way (as efficient causation), the argument above has a false presupposition. A just cannot be distinct from G here. If it is viewed as distinct, then the presupposition that A stands in (efficient) causal relation to C is false (it would be the material cause, which, on the kalam, there is no such thing[6]). If it is not viewed as distinct, then the argument has faulty premises (or at least, some premises are trivially true and the final conclusion does not follow). Because G is the efficient cause, A is not such a cause in the analysis of causation. If one insists on taking the whole thing as material causation, then the kalam not only still denies there is such a thing with respect to the initial cause and effect (God and the universe), but then the conclusion also is relatively uninteresting (i.e., God cannot be the material cause of C) and uncontroversial. This was an interesting argument, and it was fun to see where the problems lie!

[1] This also presumes there are not two causes, at least not in the same sense.

[2] Andrea Falcon, “Aristotle on Causality,” in Stanford Encyclopedia of Philosophy,, section 2, accessed April 5, 2014.

[3] Ibid.

[4] William Lane Craig, “Causal Premiss of the Kalam Argument,”, accessed April 5, 2014.

[5] This point is not trivial. This is because if the argument is interpreted this way, then it does not function as an argument against the kalam, but rather, an argument against theological determinism. This is not a problem for the indeterminist.

[6] Craig.


  1. Thanks again Randy! You know, I would love to see you discuss this on the RF forums! :D

    1. You never know...I might get back there one day. Is that thread still going?

  2. Very nice analysis Randy.

    The thread is still running:

    There are also some answers to your approach.

    1. Thanks! It doesn't appear the argument was actually addressed, sadly. The main problem is the issue of causation. They can't claim A is the material cause, since then, the kalam defender can simply deny that there are such things with respect to the kalam. If he claims A is the efficient cause, I refer him to this paragraph: "Taken this way (as efficient causation), the argument above has a false presupposition. A just cannot be distinct from G here. If it is viewed as distinct, then the presupposition that A stands in (efficient) causal relation to C is false." This is just definitional of what it means to be an efficient cause. Of course, if A is identical to G, and not distinct, and viewed as the efficient cause, then the final conclusion is invalidly drawn.

    2. It also occurred to me I might want to frame the issue, to show what it would take. In order to defeat this analysis of his argument, he would have to:

      P. Argue the thesis that there is no such thing as efficient causation.


      Q. Argue that A is both distinct from G and that A stands in efficient causal relations to C, while G does not plausibly cause A.

      I suspect that pursuit of (P) will be counterintuitive, and at any rate, highly difficult. I also think that (Q) will reveal that the whole thing makes no sense (i.e., is incoherent). After all, part of his defense rests on the idea that "A" is some action by God. If A is viewed as distinct from G and A is viewed as the cause of all of the things that are caused, we have an incoherence. But the funny thing about incoherences is that they don't tell you, by themselves, which part of the incoherent set to reject. I would just reject that A is the efficient cause, or that A is really distinct from G. The end. :)

    3. Problems seem to arise if A and G are not distinct. If you will refer back to the thread "a good argument against an uncaused cause", you may note an appeal that which possibility is actualized is random if G(A) is proposed to be the cause of any possibility that is actualized. How would you counter this?

    4. Thanks for the comment! I'd counter by pointing out that this seems to be a different objection than what the argument itself is. Perhaps it's supposed to be an objection to taking the kalam's causation as efficient? Perhaps it's supposed to be against my comment that, "If it is not viewed as distinct, then the argument has faulty premises (or at least, some premises are trivially true and the final conclusion does not follow)"? I don't really know what's being argued, and, truth be told, the statement itself (at least as you've restated it) is a little vague/ambiguous.

      Does it mean, "For any possibility P, if G is the efficient cause of P, then P is random"? That doesn't seem correct; in fact, it seems incoherent. So, does it mean, "For any causal actualization G undertakes, the effect P will be random"? Why should we think that? In any case, how is this directly relevant to the dialectic, and not a separate dialectic altogether?

      It seems to me that the dialectic is that the argument attempts to show that the kalam's causal order is incoherent by assuming the kalam's premises/entailments and deducing from them a contradiction (in effect, a reductio). But it seems to me the statement being made has nothing to do with this particular reductio, but may perhaps be starting a new one. I am open to being educated, as I have already pronounced my ignorance on what is even being said. I am not open to hopping from argument to argument, however. That's how discussions go quite too long and far, far afield.

    5. Yes, my reply was unclear, as it was more of a referral to the particular RF thread. I understand you not wanting to pursue a debate that arises from a different debate. I just thought I'd throw it out there because no one from RF seems to want to deal with the implications of A and G not being distinct. I won't go any further so I don't abuse the purpose of the comments section. The thread is there if you ever decide you'd like to respond to the matter. Thanks for the response and have a good one.

    6. That is incredibly considerate of you. Much respect! I may be interested, but if so, it will probably have to be after the weekend. I've got papers to do, papers to grade, so on and so forth. :) I'm certainly not afraid of a challenge, whatever that might be.


Please remember to see the comment guidelines if you are unfamiliar with them. God bless and thanks for dropping by!