Previously, I’ve written briefly about some of the cosmological arguments for God. I’d like to expand on this topic, and I’ll begin doing so in this post by analyzing the Kalam Cosmological Argument (KCA), since it is arguably the most well known version of the argument, which can be described with the following syllogism:
(1) Everything that begins to exist has a cause;
(2) The universe began to exist;
(3) The universe has a cause.
The conclusion of this argument is often expanded by theists to suggest that the cause must be supernaturally transcendent, immaterial, timeless, spaceless, and perhaps most importantly, this cause must itself be uncaused, in order to avoid the causal infinite regress implied by the KCA’s first premise.
Unfortunately this argument fails for a number of reasons. The first thing that needs to be clarified is the definitions of terms used in these premises. What is meant by “everything”, or “begins to exist”? “Everything” in this context does imply that there are more than one of these things, which means that we are referring to a set of things, indeed the set of all things in this case. The set of all things implied here apparently refers to all matter and energy in the universe, specifically the configuration of any subset of all matter and/or energy. Then we have the second element in the first premise, “begins to exist”, which would thus refer to when the configuration of some set of matter and/or energy changes to a new configuration. So we could rewrite the first premise as “any configuration of matter and/or energy that exists at time T and which didn’t exist at the time immediately prior to time T (which we could call T’), was a result of some cause”. If we want to specify how “immediately prior” T’ is to T, we could use the smallest unit of time that carries any meaning per the laws of physics which would be the Planck time (roughly 10^-43 seconds), which is the time it takes the fastest entity in the universe (light) to traverse the shortest distance in the universe (the Planck length).
Does Everything Have a Cause?
Now that we’ve more clearly defined what is meant by the first premise, we can address whether or not that premise is sound. It seems perfectly reasonable based on the nature of causality that we currently understand that there is indeed some cause that drives the changes in the configurations of sets of matter and energy that we observe in the universe, most especially in the everyday world that we observe. On a most fundamental physical level, we would typically say that the cause of these configuration changes is described as the laws of physics. Particles and waves all behave as they do, very predictably changing from one form into another based on these physical laws or consistent patterns that we’ve discovered. However, depending on the interpretation of quantum mechanics used, there may be acausal quantum processes happening, for example, as virtual particle/anti-particle pairs pop into existence without any apparent deterministic path. That is, unless there are non-local hidden variables that we are unaware of which guide/cause these events, there don’t appear to be any deterministic or causal driving forces behind certain quantum phenomena. At best, the science is inconclusive as to whether all phenomena have causes, and thus one can’t claim certainty to the first premise of the KCA. Unless we find a way to determine that quantum mechanics is entirely deterministic, we simply don’t know that matter and energy are fundamentally causally connected as are objects that we observe at much larger scales.
The bottom line here is that quantum indeterminism carries with it the possibility of acausality until proven otherwise, thus undermining premise one of the KCA with the empirical evidence found within the field of quantum physics. As such, it is entirely plausible that if the apparent quantum acausal processes are fundamental to our physical world, the universe itself may have arisen from said acausal processes, thus undermining premise two as well as the conclusion of the KCA. We can’t conclude that this is the case, but it is entirely possible and is in fact plausible given the peculiar quantum phenomena we’ve observed thus far.
As for the second premise, if we apply our clarified definition of “began to exist” introduced in the first premise to the second, then “the universe began to exist” would mean more specifically that “there was once a time (T’) when the universe didn’t exist and then at time T, the universe did exist.” This is the most obviously problematic premise, at least according to the evidence we’ve found within cosmology. The Big Bang Theory as most people are familiar with, which is the prevailing cosmological model for the earliest known moment of the universe, implies that spacetime itself had it’s earliest moment roughly 13.8 billion years ago, and continued to expand and transform over 13.8 billion years until reaching the state that we see it in today. Many theists try to use this as evidence for the universe being created by God. However, since time itself was non-existent prior to the Big Bang, it is not sensible to speak of any creation event happening prior to this moment, since there was no time for such an event to happen within. This presents a big problem for the second premise in the KCA, because in order for the universe to “begin to exist”, it is implied that there was a time prior in which it didn’t exist, and this goes against the Big Bang model in which time never existed prior to that point.
Is Simultaneous Causation Tenable?
One way that theologians and some philosophers have attempted to circumvent this problem is to invoke the concept of simultaneous causation, that is, that (at least some) causes and effects can happen simultaneously. Thus, if the cause of the universe happened at the same time as the effect (the Big Bang), then the cause of the universe (possibly “creation”) did happen in time, and thus the problem is said to be circumvented.
The concept of simultaneous causation has been proposed for some time by philosophers, most notably Immanuel Kant and others since. However, there are a few problems with simultaneous causation that I’ll point out briefly. For one, there don’t appear to be any actual examples in our universe of simultaneous causation occurring. Kant did propose what he believed to be a couple examples of simultaneous causation to support the idea. One example he gave was a scenario where the effect of a heated room supposedly occurs simultaneously with a fire in a fireplace that caused it. Unfortunately, this example fails, because it actually takes time for thermal energy to make its way from the fire in the fireplace to any air molecules in the room (even those that are closest to the fire). As combustion is occurring and oxygen is combining with hydrocarbon fuels in the wood to produce carbon dioxide and a lot of heat, that heat takes time to propagate. As the carbon dioxide is being formed, and the molecule is assuming an energetically favorable state, there is still a lag between this event and any heat given off to nearby molecules in the room. In fact, no physical processes can occur faster than the speed of light by the principles of Relativity, so this refutes any other example analogous to this one. The fastest way a fire can propagate heat is through radiation (as opposed to conduction or convection), and we know that the propagation of radiation is limited by the speed of light. Even pulling a solid object causes it to stretch (at least temporarily) so the end of the object farthest away from where it is being pulled will actually remain at rest for a short time while the other end of the object is first pulled in a particular direction. It isn’t until a short time lag, that the rest of the object “catches up” with the end being pulled, so even with mechanical processes involving solid materials, we never see instantaneous speeds of causal interactions.
Another example Kant gave was one in which a lead ball lies on a cushion and simultaneously causes the effect of an indentation or “hollow” in the cushion. Again, in order for the ball to cause a dent in the cushion in the first place it had to be moved into the cushion which took some finite amount of time. Likewise with the previous example, Relativity prevents any simultaneous causation of this sort. We can see this by noting that at the molecular level, as the electron orbitals from the lead ball approach those of the cushion, the change in the strength of the electric field between the electron orbitals of the two objects can’t travel faster than the speed of light, and thus as the ball moves toward the cushion and eventually “touches” it, the increased strength of the repulsion takes some amount of time to be realized.
One last example I’ve seen given by defenders of simultaneous causation is that of a man sitting down, thus forming a lap. That is, as the man sits down, and his knees bend, a lap is created in the process, and we’re told that the man sitting down is the cause and the formation of the lap is the simultaneous effect. Unfortunately, this example also fails because the man sitting down and the lap being formed are really nothing more than two different descriptions of the same event. One could say that the man formed a lap, or one could say that the man sat down. Clearly the intentions behind the man were most likely to sit down rather than to form a lap, but nevertheless forming a lap was incidental in the process of sitting down. Both are describing different aspects of the same event, and thus there aren’t two distinct causal relatum in this example. In the previous examples mentioned (the fire and heated room or ball denting a cushion), if there are states described that occur simultaneously even after taking Relativity into account, they can likewise be shown to be merely two different aspects or descriptions of the same event. Even if we could grant that simultaneous causation were possible (which so far, we haven’t seen any defensible examples in the real world), how can we assign causal priority to determine which was the cause and which was the effect? In terms of the KCA, one could ask, if the cause (C) of the universe occurred at the same time as the effect (E) or existence of the universe, how could one determine if C caused E rather than the other way around? One has to employ circular argumentation in order to do so, by invoking other metaphysical assumptions in the terms that are being defined which simply begs the question.
Set Theory & Causal Relations
Another problem with the second premise of the KCA is that even if we ignore the cosmological models that refute it, and even ignore the problematic concept of simultaneous causation altogether, there is an implicit assumption that the causal properties of the “things” in the universe also apply to the universe as a whole. This is fallacious because one can’t assume that the properties of members of a set or system necessarily apply to the system or entire set as a whole. Much work has been done within set theory to show that this is the case, and thus while some properties of the members or subsets of a system can apply to the whole system, not all properties necessarily do (in fact some properties applying to both members of a set and to the set as a whole can lead to logical contradictions or paradoxes). One of the properties that is being misapplied here involves the concept of “things” in general. If we try to consider the universe as a “thing” we can see how this is problematic by noting that we seem to define and conceptualize “things” with causal properties as entities or objects that are located in time and space (that’s an ontology that I think is pretty basic and universal). However, the universe as a whole is the entirety of space and time (i.e. spacetime), and thus the universe as a whole contains all space and time, and thus can’t itself (as a whole) be located in space or time.
Since the universe appears to be composed of all the things we know about, one might say that the universe is located within “nothing” at all, if that’s at all intelligible to think of. Either way, the universe as a whole doesn’t appear to be located in time or space, and thus it isn’t located anywhere at all. Thus, it technically isn’t a “thing” at all, or at the very least, it is not a thing that has any causal properties of its own, since it isn’t located in time or space in order to have causal relations with other things. Even if one insists on calling it a thing, despite the problems listed here, we are still left with the problem that we can’t assume that causal principles found within the universe apply to the universe as a whole. So for a number of reasons, premise two of the KCA fails. Since both premises fail for a number of reasons, the conclusion no longer follows. So even if the universe does in fact have a cause, in some way unknown to us, the KCA doesn’t successfully support such a claim with its premises.
Is the Kalam Circular?
Yet another problem that Dan Barker and others have pointed out involves the language used in the first premise of the KCA. The clause, “everything that begins to exist”, implies that reality can be divided into two sets: items that begin to exist (BE) and items that do not begin to exist (NBE). In order for the KCA to work in arguing for God’s existence, the NBE set can’t be empty. Even more importantly, it must accommodate more than one item to avoid simply being a synonym for God, for if God is the only object or item within NBE, then the premise “everything that begins to exist has a cause” is equivalent to “everything except God has a cause”. This simply puts God into the definition of the premise of the argument that is supposed to be used to prove God’s existence, and thus would simply beg the question. It should be noted that just because the NBE set must accommodate more than one possible item, this doesn’t entail that the NBE set must contain more than one item. This specific problem with the KCA could be resolved if one could first show that there are multiple possible NBE candidates, followed by showing that of the multiple possible candidates within NBE, only one candidate is valid, and finally by showing that this candidate is in fact some personal creator, i.e., God. If it can’t be shown that NBE can accommodate more than one item, then the argument is circular. Moreover, if the only candidate for NBE is God, then the second premise “The universe began to exist” simply reduces to “The universe is not God”, which simply assumes what the argument is trying to prove. Thus if the NBE set is simply synonymous with God, then the Kalam can be reduced to:
(1) Everything except God has a cause;
(2) The universe is not God;
(3) The universe has a cause.
As we can see, this syllogism is perfectly logical (though the conclusion only follows if the premises are true which is open to debate), but this syllogism is entirely useless as an argument for God’s existence. Furthermore, regarding the NBE set, one must ask, where do theists obtain the idea that this NBE set exists? That is, by what observations and/or arguments is the possibility of beginningless objects justified? We don’t find any such observations in science, although it is certainly possible that the universe itself never began (we just don’t have observations to support this, at least, not at this time) and the concept of a “beginningless universe” is in fact entirely consistent with many eternal cosmological models that have been proposed, in which case the KCA would still be invalidated by refuting premise two in yet another way. Other than the universe itself potentially being an NBE (which is plausible, though not empirically demonstrated as of yet), there don’t appear to be any other possible NBEs, and there don’t appear to be any observations and/or arguments to justify proposing that any NBEs exist at all (other than perhaps the universe itself, which would be consistent with the law of conservation of mass and energy and/or the Quantum Eternity Theorem).
The KCA Fails
As we can see, the Kalam Cosmological Argument fails for a number of reasons, and thus is unsuccessful in arguing for the existence of God. Thus, even though it may very well be the case that some god exists and did in fact create the universe, the KCA fails to support such a claim.
Here’s an excellent debate between the cosmologist Sean Carroll and the Christian apologist William Lane Craig which illustrates some of the problems with the KCA, specifically in terms of evidence found within cosmology (or lack thereof). It goes without saying that Carroll won the debate by far, though he could certainly have raised more points in his rebuttals than he did. Nevertheless, it was entertaining and a nice civil debate with good points presented on both sides. Here’s another link to Carroll’s post debate reflections on his blog.