(Also known as the argument from contingency)
It's called the Leibnizian argument because it's a modified version of an argument developed by Gottfried Wilhelm Leibniz.
The Argument
- Everything that exists has an explanation for its existence, either in the necessity of its own nature or in an external cause.
- The universe exists.
- Therefore, the universe has an explanation for its existence, either in the necessity of its own nature or in an external cause.
- The explanation of the universe is not the necessity of its own nature.
- Therefore, the universe is explained by an external cause.
Assessment
Premise (1) seems very plausible to me. It involves a restricted version of the principle of sufficient reason, according to which anything that exists either exists necessarily (meaning that its nonexistence is metaphysically impossible) or contingently (meaning that its nonexistence is possible), in which case its existence can only be accounted for in terms of the cause or causes that brought it into being. A contingent thing that has no cause would seem to be impossible.
The principle of sufficient reason implies that, for anything that exists, we can reasonably ask why it exists. We could even reasonably ask, why is there something rather than nothing? This question would only be unreasonable if things could just exist in a way that had no explanation. But this seems intuitively impossible, and to say of anything that exists that "it just does" is not a rational answer, since it would undercut scientific and philosophical reasoning. As a result, there are no rational grounds for claiming that anything could just exist for no reason. It is nothing more than a statement of blind faith.
I should mention here that many atheists have indeed claimed that the universe just exists, and that there is no explanation or that it's not sensible to look for a further explanation. I think Sean Carroll makes this claim (I am planning to dive into some of Carroll's work eventually so I will confirm if this is right), and I know for sure that John Loftus made this claim in his book Why I Became an Atheist (I think he changed the title in more recent editions). He said that the universe just exists, period. This sort of thinking is one of the main reasons I was never drawn to atheism.
Premise (2) seems pretty straightforward, but this actually turns out to be the most problematic premise in the whole argument, in my opinion. It's not because the universe doesn't exist, but because of what the argument needs to mean by "universe" in order to work. I will come back to this in a moment.
Premise (4) denies that the universe exists necessarily. In other words, its nonexistence is possible. I don't see how we could reject this premise. Otherwise I would have to think that the universe couldn't possibly have been different than it is, and this flies in the face of everything we know about the Big Bang and fine-tuning. It would also seem to require us to reject free will, since, if the universe exists necessarily, then it cannot be different than it is, and this entails that all of the events it contains could not have been different either. But is it really impossible that I could have chosen a different breakfast this morning?
If the premises are all true, it follows that the universe has an external cause. But what can we say about the cause itself? This is where the problem with premise (2) reveals itself. As a theist, I am obviously not uncomfortable with the idea that the universe has a cause. But there does seem to be a problem here. Suppose we define "universe" as the entirety of contingent reality, including all spatiotemporal reality. This is important because we want to include all of physical reality (including the multiverse, if it exists) and whatever non-physical components of reality that may exist in our world, such as our minds (if our minds exist, as I believe they do).
It follows that if the universe has an external cause, that cause cannot be a contingent thing, otherwise it would be part of the universe. So the cause must exist necessarily. And the cause of all spatiotemporal reality must transcend space and time. The only possible candidates that we know of for immaterial, timeless beings are abstract objects (like numbers), and minds. However, abstract objects don't have causal powers (that is part of what it means to be abstract; the number 8 can't cause anything). So the only known candidate for a necessarily existing cause would be an immaterial mind.
One possible objection here, then, is that the argument just assumes that the whole of contingent reality exists together as sort of a distinct object, which seems to imply a tenseless theory of time. But we seem to run into problems whether we reject the tenseless view or not.
If we reject the tenseless theory of time, then later states of the universe could potentially be explained by appealing to earlier states of the universe. Why does the universe exist at 1:30? Because it already existed at 1:29, and nothing has destroyed it since then. In this case, we could still ask why the universe exists, but we would only mean, "Where did it come from? How did it get started?" So the question of why the universe exists would still make sense, and it would still be rational to seek an explanation, but we would no longer be treating every contingent thing that exists (present tense) as part of a big, all-encompassing tenseless object whose existence needs to be (tenselessly) explained. Rather, we would be looking for an explanation of what caused it to come into existence in the first place. Unfortunately, that only works if the universe really came into existence, which the argument does not prove. If the universe (by which we still mean all of contingent reality) never had a beginning, then we could apparently explain the existence of the universe at any time in terms of its existence at earlier times, ad infinitum.
Now, in this case, it might still be reasonable to ask why a whole series of contingent causes exists, but it's not clear to me if it makes sense to treat the whole series as one distinct object. And even though it does seem reasonable to ask if the series could possibly have an infinite number of contingent causes, that would bring us to a completely different version of the cosmological argument: the Kalam cosmological argument. In order for the Leibnizian argument to work on a tensed view of time without turning into the Kalam argument, we would need to know why a beginningless series of contingent causes would, as a whole, require an external cause. Perhaps the argument could be made, but I don't know if it can.
But what if we accept a tenseless theory of time? In that case, the whole universe (including all the events that take place in it) exists tenselessly, and it makes sense to think of it as one big object. But the problem for me is that I'm not sure how to make sense of this ramification of the tenseless view of time. On the one hand, it still seems wrong to think that the whole universe-object could exist necessarily. But on the other hand, there would be no sense in which the universe has ever failed to exist!
In fact, forget the universe for a moment. Let me use myself as an example instead. I myself am a contingent object. On a tenseless view of time, it is still true that there are times at which I don't exist, times preceding my birth and times following my death (assuming no afterlife). So in that sense, I am not like the universe, since the universe includes all of temporal reality (on our definition), so there is no time at which it doesn't exist. That would be like saying that there is a time at which time itself doesn't exist. But even though I am different from the universe in that respect, I am similar to it in another respect, at least if the tenseless view of time is correct. On a tenseless view, it's not as if I ever "came into existence" or will ever "cease to exist," because these concepts require a tensed view of past and future. On the tenseless view, I exist tenselessly, and so does the whole temporal world. I am just a small part of the universe that exists tenselessly. It never really begins or ends, and neither do I.
One of the conceptual problems I have with tenseless existence is that I don't know how to draw a meaningful distinction between tenselessness and timelessness. Obviously, we can specify different sections of the tenseless universe and say, "This part is earlier than that part." But our choice of which direction is earlier or later seems arbitrary. It seems that we are smuggling in an awareness of temporal order that stems from our psychological experience of the passage of time, which is just an illusion if the tenseless view of time is correct. All of which is to say, it is hard for me to understand the difference between timelessness and tenselessness, so if we speak of a tenseless universe being "caused" by a timeless creator, I truly don't know what that means.
There is a semantics that philosophers use when talking about necessity and possibility involving possible worlds. A contingent thing only exists in some possible worlds; a necessary thing exists in all possible worlds; an impossible thing doesn't exist in any possible world. So if the universe is contingent and its cause is necessary, then there are some possible worlds where the cause exists without the universe. (As an aside, some philosophers think it is impossible for a necessary cause to have a contingent effect, but that is because they assume that the cause cannot be a mind endowed with free will. I won't get into that here.)
The problem I have is that, since tenselessness seems indistinguishable from timelessness, then having God exist "timelessly" with an apparently timeless universe just makes it seem like the universe exists eternally, without beginning or end. In that case, I'm not sure how one can justify there being a possible world where God exists without the universe, since the universe apparently exists with God for all timeless eternity.
I don't conclude from this that the universe might exist necessarily, otherwise I would have to conclude that I exist necessarily, and that there is no free will, since I couldn't possibly have done differently than I did. All of that seems absurd, so I am happy to continue to think that the universe is a contingent thing. But I just cannot make sense of thinking about any of this in tenseless terms. This is not a failing of the Leibnizian argument, but (I think) of the tenseless view of time. I just don't think the tenseless view of time makes sense. But then, I don't have much interest in an argument that requires a tenseless view of time in order to work.
Incidentally, if my analysis here is incorrect, and there is a way to preserve the contingency of the universe even on a tenseless view, then I think the above version of the Leibnizian argument actually does go through successfully. But I am just not sure about the tenseless, contingent universe.
This doesn't mean that I see no value in the Leibnizian argument. But its value for me is in establishing the necessity and personhood of an uncaused first cause. It would go something like this:
- Everything that exists has an explanation for its existence, either in the necessity of its own nature or in an external cause.
- There is a first cause in the series of contingent causes.
- Therefore, the first contingent cause has an explanation for its existence.
- The explanation of the first contingent cause cannot be the necessity of its own nature.
- Therefore, the explanation of the first contingent cause is an external cause.
Premise (1) is the same as before. Premise (2) presupposes a series of contingent causes, which seems unobjectionable to me, though it undoubtedly raises questions about the nature of causation that I don't have any desire to explore here. To say that there is a first contingent cause in the series is obviously controversial, but I would accept it on similar grounds as those that come into play in discussion of the Kalam cosmological argument. I've even argued for metaphysical finitism elsewhere on the blog, so I'll just assume that the series of contingent causes cannot go back forever.
It is possible, of course, that there could be multiple contingent causes at the beginning of the series, especially if there are multiple series of causes that are causally unconnected to each other. These would have to be either timelessly existing causes, or temporal causes that exist simultaneously. As I've already indicated, I am not sure if I can understand the idea that something exists timelessly but contingently. I am inclined to think that contingent things can't be timeless. But whether the first contingent causes were timeless or temporal, the group itself would be a contingent thing whose existence required an explanation in the form of an external cause.
Premise (4) is true by definition.
We are left with an uncaused first cause that must exist by a necessity of its own nature, and it cannot be an abstract object, nor can it be spatial or temporal, so we end up with a necessarily existing immaterial mind, just as before.