I stumbled upon an argument written by Richard Carrier, who I am now considerably less impressed with than I was prior to running into the argument. When first I read through it, I thought “wow, this argument is remarkably bad”, and that first impression has been reinforced as I’ve tried to take the argument seriously. Here it is, at least as I found it on this blog. The argument is intended to be a response to the Kalam cosmological argument for the existence of God, and thus is an attempt to explain that the universe could come into being without God being the cause of the universe’s coming into being.
- P1: In the beginning, there was absolutely nothing.
- P2: If there was absolutely nothing, then (apart from logical necessity) nothing existed to prevent anything from happening or to make any one thing happening more likely than any other thing.
- C1: Therefore, in the beginning, nothing existed to prevent anything from happening or to make any one thing happening more likely than any other thing.
- P3: Of all the logically possible things that can happen when nothing exists to prevent them from happening, continuing to be nothing is one thing, one universe popping into existence is another thing, two universes popping into existence is yet another thing, and so on all the way to infinitely many universes popping into existence, and likewise for every cardinality of infinity, and every configuration of universes.
- C2: Therefore*, continuing to be nothing was no more likely than one universe popping into existence, which was no more likely than two universes popping into existence, which was no more likely than infinitely many universes popping into existence, which was no more likely than any other particular number or cardinality of universes popping into existence.
- P4: If each outcome (0 universes, 1 universe, 2 universes, etc. all the way toaleph -0 universes, aleph -1 universes, etc. [note that there is more than one infinity in this sequence]) is no more likely than the next, then the probability of any finite number of universes (including zero universes) or less having popped into existence is infinitely close to zero, and the probability of some infinite number of universes having popped into existence is infinitely close to one hundred percent.
- C3: Therefore, the probability of some infinite number of universes having popped into existence is infinitely close to one hundred percent.
- P5: If there are infinitely many universes, and our universe has a nonzero probability of existing (as by existing it proves it does, via cogito ergo sum ), then the probability that our universe would exist is infinitely close to one hundred percent (because any nonzero probability approaches one hundred percent as the number of selections approaches infinity, via the law of large numbers ).
- C4: Therefore, if in the beginning there was absolutely nothing, then the probability that our universe would exist is infinitely close to one hundred percent.
It is difficult to read this without immediately feeling as though it is nothing other than the most indulgent sophistry gone awry, but upon closer and more charitable reading it seems worse than mere sophistry; it seems to be altogether confused. Let alone the fact that it abjectly rejects any robust expression of the principle of sufficient reason (PSR) – it seems as though it even goes as far as assuming the falsity of Parmenides’ ex nihilo nihil fit (out of nothing nothing can come). This principle seems to be self-evident to me, and thus P3 is self-evidently false, and it is at that point that the argument goes entirely wrong. This third premise completely denies the PSR, and by reason of that alone seems to defy the canons of rational thought.
We could end our treatment of it there, but since Carrier’s modal intuitions apparently do not obviate a conflict between the falsity of the PSR or Parmenides’ principle and rational thought, I think maybe the job of a charitable apologist is to go further. Let us assume for the sake of argument that ~PSR, or at least let us assume that the truth value of the PSR is inscrutable, and adopt an apologetic attitude of dialectical charity. What other feature of his argument goes wrong (of course, if one gets the PSR wrong, then I have to wonder what they can possibly get right, but again, let’s try to be gratuitously charitable). The crux of Carrier’s argument is this notion that the likelihood of our universe popping into existence uncaused from absolutely nothing is infinitely close to 100%. I think it is at this point, in his fifth premise, that we can find impetus for an interesting objection; since there are infinitely many universes which have a probability infinitely close to 100% of being actualized, the question for an epistemically responsible person should become what kind of universe do we occupy? Let me appeal to an objection William Lane Craig brings up to the Multiverse-hypothesis being used as a response to the Teleological argument from fine tuning: he demonstrates if the Multiverse hypothesis is true, then not only are atheists not justified in believing in things like evolution, but they aren’t even justified in believing in the external world. Craig makes this argument by appealing to the notion of ‘Boltzmann’ brains – which are essentially brains constituting universes on their own, and which also ‘perceive’ an external world merely by accident. He suggests (rightly) that these universes are simpler than the models we currently entertain of our universe, and are therefore more common among Multiverse universes, and thus our universe is more likely a Boltzmann-brain universe than a universe like the one we imagine ourselves to exist in if the Multiverse hypothesis is true. However, that entails that the most likely universes which could pop into being uncaused from nothing and in which we are (or ‘I’ am) observer(s) are Boltzmann brain universes. If this thinking is correct then although it is the case that what we take to be our universe, call it U1, has a probability infinitely close to 100% of popping into being uncaused from nothing, any Boltzmann brain universe, call it Ub, also has a probability infinitely close to 100% of popping into being uncaused from nothing. Moreover, a Ub has a probability infinitely closer to 100% than the probability of a U1 of being actualized because it is simpler. It therefore seems that the naturalist who accepts Carrier’s reasoning should be committed to believing in a Ub rather than a U1, on pain of having maintaining beliefs irrationally or in an epistemically unjustified way.
One might respond to this by saying that since Carrier isn’t committing himself to a multiverse hypothesis necessarily, it seems as though neither Ub nor U1 is any more likely prima facie than the other (each other) or any other universe – complexity, one could argue, does not make something less likely to pop into being uncaused. I don’t think that response takes the seriousness of the objection into account, since a well recognized empirical-methodological rule which governs the epistemology of all or most Naturalists is Occam’s Razor (since Naturalists have to be empiricists, and cannot be rationalists – all rationalists in the history of philosophy have been Theists, and most precisely because of the PSR). Quite apart from whether it is more likely that a simple universe pop into existence uncaused from nothing than that a more complex universe do the same (which seems evident to me, but perhaps Carrier’s intuition would lead him to disagree) – a Ub ought to be epistemically preferred simply because it is a simpler explanation, and equally ‘sufficient’ insofar as it explains satisfactorily all perceptions of the external world (at least satisfactory for the person who rejects the PSR). Moreover, since a multiverse could also pop into existence uncaused from absolutely nothing, and since complexity has no bearing on probability, it seems as though U1, Ub and let’s call it Um (for multiverse) are all equally probable. It then seems as though of the three possibilities U1, Ub and Um, the probability P of it being the case that we should conclude that we occupy a Boltzmann brain universe is at least two thirds (P ≥ 2/3). There are also other universes which may instantiate and which are relevantly dissimilar to U1 in such a way that it would be equally absurd to believe in them – it would literally be to abandon reason, and perhaps even rational thought altogether, to commit oneself to believing in them.
Interestingly, Carrier apparently wanted to obviate the falsity of the PSR by making the following follow up argument: “that something cannot come from nothing is a law, which is something. Thus, if we have nothing, we don’t have that law, meaning something CAN come from nothing.” This is particularly poor thinking. For a Naturalist to say that a principle of logic is ‘something’ is confused because, as I noted earlier, the Naturalist is bound to be an empiricist, and a consistent empiricist will think of laws as useful models rather than anything like platonic forms. Moreover, if the Naturalist even admits that the laws of logic can possibly exist as anything like platonic forms then they have fallen into a pretty dangerous trap of having already implicitly conceded the existence of God, since if it is even possible for laws of logic to exist independent of cognitive models, then the fact that the laws of logic are ‘necessary’ thoughts, which require a necessary mind and thus a necessary being, together with the modal ontological argument which, with the use of axiom S5 of modal logic, has demonstrated that if it is even possible for a necessary being to exist then a necessary being does exist, logically entails the existence of a necessary being in whose mind reside the laws of logic – therefore it would be necessary to concede the existence of God. Finally, even if laws did exist as platonic forms it seems uncontroversial to maintain that such things would be causally effete. Worse still, to say that something ‘can’ come from nothing would itself be a ‘law’, and since it doesn’t exist if nothing exists, it would not be possible for something to come from nothing (again though, the whole conversation is fundamentally confused because of the assumptions at play being self-evidently false, such as that the state at which nothing exists can possibly give rise to the state in which everything exists).
For the sake of rhetoric, we can put aside the objection that Carrier is obstinately disregarding the Principle of Sufficient Reason, and respond instead by showing him that even if we accepted his own logic it would be infinitely more likely that Boltzmann-brain-ian solipsism is true in this universe than that the universe the ‘I’ of the cogito ergo sum observes is not illusory. If any universe can pop into being uncaused, then the likelihood, given that ‘I am perceiving’, of our universe fitting the description of a U1 is infinitely less likely than our universe fitting the description of a Ub. That suggests that the perception of this universe comes about by brute perceptions (perceptions which are brute facts without any sufficient explanation). Thus by this reductio ad absurdum we can demonstrate that Carrier’s reasoning ought to lead him into positions which are even more uncomfortable than Theism (though if he was willing to sacrifice principles of reason like the PSR just in order to avoid believing in God then who knows, maybe he’ll be comfortable with brute solipsism). We might be able to point out another uncomfortable feature of this brute solipsism is that it makes it literally impossible to have justified beliefs about anything (other than that the ‘I’ exists), and any world view which makes having reliable beliefs impossible or nearly impossible is epistemically self-defeating in such a way that one would have to be irrational to accept it. Perhaps Carrier would still prefer this to Theism though – but at that point what exactly can a reasonable apologist do?
A final thought, if I were an Atheist who agreed with Carrier’s argument, and was faced with the objection I just raised, I would likely opt for accepting something like Alvin Plantinga’s reformed epistemology, and simply argue that belief in U1 is properly basic – but upon further reflection that would be an inadequate response. Carrier’s argument itself proposes that any universe can itself come into existence uncaused from nothing, and thus calls into question the legitimacy of reformed epistemology in this case since it offers a defeater to the belief in U1. It is not merely the case that the probability of U1 is inscrutable, but rather according to Carrier’s presuppositions it seems as though U1 is infinitely less probable than a host of other options which themselves make belief in U1 unjustified and untenable.