Wednesday, August 24, 2011

The Kalām Cosmological Argument


It’s been difficult deciding on an initial topic. Most lay-level topics have been more or less sufficiently dealt with at numerous points and by numerous internet authors. For instance, the point at which I’ve decided to begin is a family of arguments for the existence of God known as “cosmological arguments.” These are arguments essentially for a first cause of the universe, and they take many forms. It is important to remember that a defeater for one argument will not necessarily be a defeater for another, even if both are cosmological arguments.

To begin I’d like to discuss the “kalām cosmological argument” (KCA) which has its origins in medieval Christian and Muslim philosophy. It has taken various forms in the many centuries since its formation, but the most prominent modern formulation is articulated this way by its most well-known contemporary defender, Christian philosopher and theologian William Lane Craig:

Premise 1: Everything that begins to exist has a cause.
Premise 2: The universe began to exist.
Conclusion: Therefore the universe has a cause.

This is what’s known in logic as a categorical syllogism, and it is in the AAA-1 form. It is unconditionally logically valid. This is to say that there are no formal logical fallacies committed by the argument: if each premise is true, then the conclusion is necessarily (not merely probably) true as well. It may seem trivially true that the universe has a cause, and the temptation may arise to ask in what way this can be construed as an argument for God’s existence. Is God simply identical with a cause of the universe? This proclivity is, strictly speaking, correct. The argument does not conclude that God exists. Rather it concludes that there is a cause of the universe. Here is it necessary to say that the universe is generally taken to mean all matter and energy, all space and time. And from this definition it follows that if there is a cause of the universe, the cause cannot be a consequence of or identical to matter or energy, to space or time. The cause of the universe is a timeless, immaterial, unextended entity. To suggest that the universe is caused by a prior material state is to commit oneself to logical contradiction, for there is no state which is “prior” to time, for “prior” is a temporal word. Likewise there is no material state which could serve to cause all material reality to come into existence. This would denote a material existence before any material, which is to postulate both “It is the case that A” and “It is not the case that A,” which is logically contradictory and therefore logically impossible. So as a consequence of what is meant by the term “the universe” it follows that the cause of such a reality must be something other than that which constitutes the universe.

Another significant question now arises because it may be initially unclear why the cause of the universe should be thought of as a personal being rather than an impersonal reality of some kind. In the first place, it must minimally be admitted that the cause of the universe is an efficient cause; the cause of the universe is that which causes the universe to begin to exist. It is difficult to imagine an actual non-personal entity which could have productive capacity of this kind. It is furthermore difficult to see how such an entity could be timelessly actual and yet could bring a temporal universe into existence a finite time ago. That is to say, without time, there is no change. So whatever else the cause of the universe could be, it is a changeless entity, at least without the universe. It seems essential then to conceive of the cause of the universe as an unchanging reality, and yet it is also essential to conceive of it as an entity with productive capacity. This gives us strong reason to resist seeing the cause of the universe as an impersonal force of some kind, since we might ask “Why did the universe begin to exist 13.7 billion years ago rather than 14.7?” If the cause of the universe is a sufficient cause, then for every point at which the sufficient cause is present, the universe will also be present. Yet the universe is temporally limited; it began to exist. This strongly disconfirms the impersonality of the cause of the universe.

But what other option remains? I suggest that a personal cause of the universe which possesses free will is a more plausible explanation, for such an entity could exist timelessly (as the cause of the universe would) and yet could cause the universe at a specific point a finite time ago by an exercise of its free will. This seems to be the only plausible account of how an a-temporal reality could cause a temporal reality to begin to exist.
The strength of the kalām cosmological argument can be seen best by considering some objections to it. I would not waste my time by mentioning some of the (in my estimation) worst objections, such as the insistence that if everything has a cause of its existence, so must God. Any serious objection must attack one or the other premise of the argument, or both. Consequently only those objections which do this will be considered.
First it is objected (a) that quantum mechanics gives us reason to doubt the first premise. It is asserted that we have examples of things beginning to exist without causes, and hence premise one is false. Alternately, it is alleged (b) that the proponent of the KCA cannot demonstrate premise one to be true, because to do so would be to prove a universal affirmative, which, like proving a universal negative, is notoriously difficult to do.
With respect to (a), the examples cited from quantum mechanics do not in fact truly represent the coming-into-existence-from-nothingness-without-a-cause of something. Rather they seem to describe indeterminate causes rather than uncaused geneses. Something indeterminately caused is nevertheless caused. Second, the examples cited from quantum mechanics are not proven counter-examples, since all that can be said, at best, of these seem to be that we have hitherto found no cause. This is clearly distinct from the allegation that we know definitively that there is no cause.

Moreover we have strong reason to affirm premise one of the argument, since to deny it is to suggest that at least one thing has come into existence (from nothing) without a cause. Once it is accepted that nothingness has some productive potential, it becomes impossible to explain why we do not observe things coming into existence uncaused from nothingness more frequently. For instance, Craig notes that we cannot in principle explain why we do not observe bicycles and Beethoven and horses, et. al. popping into existence before our very eyes. After all, it cannot be claimed that nothingness only causes universes and sub-atomic particles to begin to exist, because nothingness just has no properties and so has no restrictions on what it could “cause” to begin to exist. For these and other reasons, premise (1) seems more plausibly true than false.

With respect to (b), this seems to indicate a poor understanding of the purpose of an argument for a certain conclusion. If true, the proponent of (b) has actually shown that we are rationally justified in accepting as true only that which we can prove with one hundred percent certainty. This might include propositions which are true by definition, such as “There are no round squares,” but very little else. But surely the proponent of (b) does not wish to go this way, for if they are correct that we are rational in accepting only those truths which cannot possibly be disputed, all the opponent of (b) need do is dispute (b). It would follow from such a dispute that the proponent of (b) is not rationally justified in accepting (b). Which would dissolve her objection to the KCA.

A more generous construal of (b) might be that the usefulness of the KCA is diminished because premise one is in some sense dubitable. But the proponent of the KCA might fully agree that it is possible to reject the KCA, and nevertheless see this as no flaw or weakness of the KCA as such. After all, it is possible to reject whatever the proponent of (b) says too! Yet the proponent of (b) does not take this to mean that (b) is not rational or is rationally dubitable. It seems clear therefore that all a good argument needs to do is provide rational grounds for affirming its conclusion; it is not necessary to show that the negation of its conclusion is impossible. It must additionally be considered whether, in light of the KCA, it is plausible to think that there is no cause of the universe. This must be the aim of both the proponent and the opponent of the KCA, since to suggest otherwise is to suggest something generally abhorrent to the opponent of the KCA: that unless they can prove their objection with 100 percent certainty, it is as plausible to think they are correct as incorrect.
Objections to premise (2) of the KCA take something like the same form as those for premise (1). First it is contended (c) that the universe did not begin to exist, that the universe is in some sense eternal and therefore does not need a cause. Second, it is contended (d) that the proponent of the KCA has not shown premise (2) to be true.

In answer to (a), we might say that the most viable cosmological models to date do include a beginning of the universe. We might argue, as the medieval thinkers did, that an actually infinite number of things (events in time) cannot exist, or alternately that a set containing an actually infinite number of things (events in time) could not be formed by adding one member after another. I see each of these arguments as very plausible.

In short, the kalām cosmological argument is a deductively valid argument with very plausible premises and a conclusion, the best explanation of which is that a timeless, space-less, immaterial extremely powerful personal being exists and caused the universe to come into existence. 

12 comments:

  1. Really great job breaking everything down to accessible levels.

    ReplyDelete
  2. (P1) Everything that begins to exist has a cause.
    (P2) The universe began to exist.
    (C) Therefore the universe has a cause.

    Three things to say here.

    First, I find that the available scientific evidence does not justify any inference to statements of ultimate causation – or lack of it.

    In fact, on the balance of evidence, I find that the matter tips in favor of non-causality. In other words, causality appears to break down completely in quantum processes, and good experimental data show that quantum vacuum fluctuations and virtual particles come into existence (i.e. create measurable effects such as the Lamb effect and the Casimir-Polder force) acausally all the time.

    Thus, I reject (P1).

    Second, (P2) ought to be rejected because it assumes a scientifically implausible conception of time. If time and space came into being at the initial development of the universe, then there could be no time in which the universe 'began' to exist. Similarly, it certainly could not be the case that god could have 'caused' the universe to come into being since for y 'to cause' x entails that there are temporal relations which exist between y and x. But per the hypothesis there is no time before x, and thus no temporal relations which might exist between y and x.

    Third, regarding actual infinities. Theists cannot avail themselves of standard cosmological models of the origin of the universe *and* employ finiteness arguments consistently. E.g., in Big Bang cosmology, the infant universe is held to entail infinite density, pressure, and temperature if interpreted via general relativity (we have no other macro-level interpretation).

    That aside, there is no contradiction or absurdity entailed in saying that before this point in time there was a preceding point in time and before any point in time there was a preceding point in time. Infinite sets and sequences are permitted (under intuitionistic logic, from which Craig gleans some of his arguments against actual infinite sets) if there is a procedure which defines that set or sequence. In transfinite arithmetic, that procedure is simple counting; and likewise for time.

    ReplyDelete
  3. Daniel,

    Apologies. I really ought to address your bit about quantum events and causation.

    You employ (P1) either as a strong statistical generalization or as an a priori truth. If the latter, then (P1) is patently false. First, I would argue- pace W.V.O Quine- that there are no a priori truths. Second, there is no logical contradiction in conceiving of an event which has no cause.

    To the former, one can either let one's (antiquated medieval) metaphysics direct one's science, or let one's science direct one's metaphysics. Apparently, and unadvisedly, you insist on former: That all that which begins to exists requires a cause is very much an empirically testable hypothesis and the *best* experimental data reveal that causation breaks down in a fundamental way. To be sure, the data indicate that causation is rather a macro-level description and has no home in the quantum world.

    Of course, one could advance the argument that it is *possible* that a future standard model will reveal that quantum level phenomena admit to causal relationships, but (1) that does not warrant (P1) *now* and (2) that does make one's belief that quantum level events are causally determined rational now; to insist otherwise is to force a pre-scientific metaphysical intuition upon our best and current relevant physics.

    ReplyDelete
  4. Apologies for the incivility. Correction:

    "...and (2) that does NOT make one's belief that quantum level events are causally determined rational now;"

    P.S. Btw, Daniel, I applaud you in your efforts here at your blog. I hope that we can have fruitful discourse in the future.

    ReplyDelete
  5. Eamon,

    Thank you for your very well articulated thoughts. Occasional condescension aside, there is no incivility for which to apologize. I wonder if we could clarify your objections a bit, though.
    You object to P1 on the grounds that a) it is not justified by available scientific evidence, and b) there appear to be empirically verified counter-examples, such as a-causal geneses on the quantum level. Thus, you say P1 not only oversteps what could in principle be justified by scientific evidence, but also appears to be false.
    You object to the second premise because it assumes an account of time which you view as scientifically implausible. You say that in this case, there is no time for the universe to begin to exist, yet such a time seems necessary if the universe did indeed begin to exist. By extension, your argument seems to imply that time itself could not begin to exist, for there would be no time in which this beginning could occur. And by further extension, time must be eternal in the past (but since time is measured by relationships between events, it follows that events must also be eternal in the past in an earlier-than way). You additionally contend that God could not have (I hope I’m not phrasing your contention too strongly here) caused the universe to begin to exist, because causes precede their effects. But God could not “precede” time itself, for such an account assumes temporality. Thus the view that God (or anything for that matter) caused the universe to begin to exist seems logically incoherent.
    Next you object that theists who accept the standard models of cosmology but reject actual infinities are being inconsistent, since the standard cosmological model entails (via general relativity) infinite heat and density in the initial cosmological singularity.
    Is this a generally accurate summary of your objections? Feel free to amend as you see fit.

    ReplyDelete
  6. Daniel,

    Regarding (P1), you have (roughly) encapsulated my position well-enough. Though I should to clarify my position here just a bit. At the level of macro physical level there is an appearance of causal order, but, from the micro physical level, the appearance depends rather upon local facts about entropy, i.e. the statistical distribution of micro physical particles from low entropy to high entropy. There certainly seems to be nothing here which reveals anything which might approximate an underlying causal order. E.g. typically we want to say that the temperature of the warm glass of water caused the ice cube to melt, but instead the matter is such that a certain assemblage of particles (the ice cube) and another assemblage of particles (the warm water) possess a probability to freeze solid, dissipate into atmosphere, reach equilibrium, pass through the glass, etc. Though equilibrium is more probable, the relationship is not deterministically causal; neither is it at the fundamental level indeterministically causal; it is acausal through and through.

    That aside, regarding (P2), I would like to offer a few clarifications. You write:

    “You say that in this case, there is no time for the universe to begin to exist, yet such a time seems necessary if the universe did indeed begin to exist.”

    (Note that below there is much not mentioned [in particular Dr Craig's implicit employment of the contentious A-theory of time in the KCA], but the following remarks should suffice.)

    The best relevant cosmological models entail that space-time came into existence with the 'Big Bang'. Thus, the universe was not created 'in' time, which is to say there was no time 'before' the universe. (To assert otherwise is ipso facto to assert that there is something like a Newtonian absolute time in which the universe began to exist.) Hence, we may say that there is no time at which the universe did not exist. It follows also that the initial 'Big Bang' is not an event, since events are defined as occurring in space-time. Thus, the proposition 'The Big Bang had a cause' is incoherent / not well-defined.

    Re: “You additionally contend that God could not have (I hope I’m not phrasing your contention too strongly here) caused the universe to begin to exist, because causes precede their effects.”

    To assert 'x caused y' is to assert that there exists a temporal relationship between x and y such that x precedes y in time. But if there is no time 'before' the universe, then are (1) no temporal relationships and (2) no time in which one can meaningfully assert (a) that 'x caused y' and (b) that 'x precedes y'.

    I understand that Dr Craig has advanced the argument that causes can be simultaneous with their effects. But this argument holds little plausibility.

    First, the 'Big Bang' is not an event and thus cannot be an effect of a cause- simultaneous or otherwise.

    Second, if one asserts that there is simultaneous asymmetric causation, one must provide criteria by which to distinguish one of the two (or more) causally connected simultaneous events as the cause of the other. In nuce, one must provide criteria by which to distinguish the causal relationship which is purported to hold between x and y.

    Third, simultaneous assymetrical causation entails that the causal influences between x and y propagate instantaneously, which is problematic since the special theory of relativity limits the transmission of causal influences to the speed of light. Thus, x and y cannot occur simultaneously.

    ReplyDelete
  7. Eamon,


    Thank you for these clarifications. I ask for them because I’m planning to make your comments the subject of a future post concerning some objections to the KCA. I want to be pretty certain that I understand what your objections are, and that I don’t present a straw man in my replies to them. So anyway, I think I need a bit more by way of clarification, however. First if you could clearly define “micro physical level.” Do you mean the atomic level or perhaps the quantum level? Do you mean to say that everything which appears on a macro physical level to be causally related actually reduces to a-causal movements toward entropy? For instance, particles which begin to exist in the quantum vacuum are not really caused to come into existence as much as in some way being an expression of increasing entropy? I’m afraid I don’t understand your view very well.
    Do you mean to say that states of affairs on a micro physical level move from low to high entropy, but that this movement has no causes? Does entropy increase for a reason?
    You offer an example of a glass of warm water meeting an ice-cube, and suggest that warm water does not cause the ice cube placed within it to melt because the ice-cube has the potential to, for instance, evaporate, condense on the glass, etc… rather than to melt. You say that probably the ice-cube will melt and the warm water will be slightly cooled (equilibrium), but that this is only probably the case. I don’t see how you get from this to the assertion that it is a-causal “through and through.” Could you explain that a bit more?
    Dr. Craig explicitly rather than implicitly “employs” (I’m not sure this is the proper word) the A-theory of time. Do you see this as problematic?
    Correct, the universe was not created “in” time; rather time and the universe began to exist at the moment of the big bang. For the purposes of the KCA, of course there was no time “before” the universe; this is logically incoherent. But you move from this to: “we may say that there is no time at which the universe did not exist.” But this seems equivalent to saying that there is no time when there was no time. This seems true by definition, like saying there is no place where there is no space. But it by no means follows that the universe has existed for an infinite number of equal finite intervals. If we negate this, then we assert “it is not the case that the universe has existed for an infinite number of equal, finite intervals,” or simply, “it is not the case that the universe is eternal in the past.” Yet if it is not past-eternal, but exists now, then it began to exist a finite time ago, as even you seem to agree that the universe came into existence with the big bang. So you seem to agree that the universe began to exist. Which is what P2 asserts. Why exactly, then, do you reject P2?
    You say that events are defined as temporal. If you could defend this claim… Perhaps cite a source or two.
    Also, for my sake, you at times seem to agree that I’ve accurately summarized your view, but you go on to discuss it further. If you wouldn’t mind being a bit more clear about where I’ve got it right and where I’ve got it wrong, that would be appreciated.

    ReplyDelete
  8. I feel I should be a bit more specific.
    The words you now read have appeared on this site. They were not here yesterday. Would you agree with this?

    If so,is it your view that these words appear here a-causally? What role, if any, would you say the movements of my fingers over a keyboard, or the firing of neurons in my brain played in the appearance of these words?

    You say the common-sense view is that warm water causes the ice-cube to melt. Yet you reject the common sense view. Do you reject it on scientific grounds, or philosophical, or both? Is it your view that science demonstrates ice to melt a-causally?

    ReplyDelete
  9. Daniel,

    The comments regarding (P1) were made to clarify my position on the causal principle; I apologize for the confusion. I will attempt to re-clarify.

    In brief, the causal principle is not an a priori / metaphysical truth but rather an empirical generalization of macro-level phenomena. At the micro-level [by which I mean the atomic and subatomic, or any microscopic level at which quantum processes are considerable] causation breaks down. I merely used thermodynamics to show that the causal principle does not hold at the microscopic level. Thermodynamical states are a-causal insofar as, given a preceding state condition, subsequent states conditions obtain randomly- that is, by chance- irrespective of the initial conditions. This violates causation insofar as causation standardly entails that effects invariably follow their causes. In other words, standard accounts of causation entail that given an initial state condition [angular momentum, velocity, spin, etc.], a subsequent state condition obtains deterministically.

    The mention of quantum vacuum fluctuations in the my first comment was meant as an another counterexample to the causal principle, though in that example we see that something [i.e. quantum particles] come from nothing and behave indeterministically a-causally.

    Re: 'Dr. Craig explicitly rather than implicitly “employs” (I’m not sure this is the proper word) the A-theory of time. Do you see this as problematic?'

    Yes. An A-theory of time is no longer plausible given Einstein's special theory of relativity and thermodynamics.

    Re: ' But this seems equivalent to saying that there is no time when there was no time.'

    No. My statement is equivalent to saying that at no time did the universe not exist. Thus, your subsequent comments on this point are not germane.

    Re: 'Why exactly, then, do you reject P2?'

    Because (P2) assumes a scientifically implausible conception of time and for the reasons I stated in my previous comment.

    Re: 'You say that events are defined as temporal. If you could defend this claim… Perhaps cite a source or two.'

    I am not sure what you mean here. It seems from your comment that I may infer that you hold that events can occur a-temporally?

    Re: 'The words you now read have appeared on this site. They were not here yesterday. Would you agree with this?'

    Yes.

    Re: 'If so, is it your view that these words appear here a-causally? What role, if any, would you say the movements of my fingers over a keyboard, or the firing of neurons in my brain played in the appearance of these words?'

    As a macro-level description we could say that you were a causal influence of the words to appearing here. However, at the micro-level the appearance of causation depends rather upon local facts about entropy, i.e. the statistical distribution of micro physical particles from low entropy to high entropy. Thus, at the micro-level causation fails.

    Re: 'You say the common-sense view is that warm water causes the ice-cube to melt. Yet you reject the common sense view. Do you reject it on scientific grounds, or philosophical, or both? Is it your view that science demonstrates ice to melt a-causally?'

    I reject the common sense view insofar as one attempts to superimpose it upon fundamental systems which do not warrant the imposition. I reject the causal principle on both scientific grounds and, by entailment, on philosophical grounds. Philosophically, I am a scientific naturalist pace Quine (though, unlike Quine, I am a mereological nihilist) and therefore I do hold that philosophy (or, more better, conceptual analysis) has anything to recommend science at all.

    I should end by saying that my position on causation and time is not uncontroversial amongst physicists. Also, I should say I do not reject causation en bloc; causation is a primary generalization on macroscopic systems. I simply find no reason to believe that the causal principle holds in the microscopic world.

    ReplyDelete
  10. Correction:

    "... therefore I do NOT hold that philosophy (or, more better, conceptual analysis) has anything to recommend science at all."

    ReplyDelete
  11. Thank you; these clarifications do help a great deal. Not let me clarify:

    You stated that events are defined as temporal, and I asked you to cite a source or two which so define the word "event." You ask whether I'm suggesting the possibility of a non-temporal event; the answer is no. I'm questioning the truth of the word "event" being so defined.

    You suggest that the beginning of the universe as such cannot then be an event, since events happen temporally, but the beginning of the universe was also the beginning of time itself; this seems to me a bit confused. The big bang, since it entails change and therefore time, does take place temporally, and is therefore an event, under your characterization of an event. How then do you assert that it is not an event?


    Expect a post-length reply sometime in the next couple of months. My wife and I are moving and having another child, so I expect to be quite busy with that for some time.

    ReplyDelete
  12. Daniel,

    First, congratulations on the baby; I hope all goes well.

    Now, events, as commonly conceived, are physical states of affairs which occur in space-time. That is, events, if they are to have a physical interpretation, must be assigned three spatial co-ordinates and one temporal co-ordinate within a specified frame of reference.

    For a reference, see Albert Einstein's own 'Relativity: The Special and General Theory' (Wings Books, NYC) pp. 105 - 107. For a more updated reference, see James Hartle's 'Gravity: An Introduction to Einstein's General Relativity' (Addison-Wesley).

    ReplyDelete