After an exchange of articles with Andrew Loke (see Why Nothing Remains a Problem: The Andrew Loke Fiasco and his reply) we were asked by a private professional forum to debate the topic of the Causal Principle, which anchors the first premise of the Kalam Cosmological Argument (or KCA), to wit, “Everything that begins to exist has a cause.” Because Loke tried to take me to task for my Argument from Nothing approach against Kalam-like premises, I took that as the critical approach in our debate (see The Problem with Nothing: Why The Indefensibility of Ex Nihilo Nihil Goes Wrong for Theists and Koons Cosmology vs. The Problem with Nothing). A video of that debate (along with Q&A from some notable philosophers) may be posted publicly someday (I have no control over that). If it does, I will link it here as well. But here’s my perspective on it.

Summary of the Debate

The debate was a rout. Loke completely failed to defend his central proposition (and indeed, it was central by his own admission) that if the causal principle had ever failed to operate, then we should still observe its failure to operate now. I demonstrated the opposite is the case: it will fail to operate only when no entity exists to manifest it; and we are not in that condition anymore: even empty spacetime is a substantive, seething ‘something’ generating the causal principle. Loke completely failed to explain how a causal principle can exist when nothing exists—as then not even principles will exist, so he was contradicting himself. And he completely failed to explain how, if the universe began to exist (and nothing else existed at the time), that universe even could have had a cause, much less had to have had one. Causes cannot precede themselves.

This is the fundamental problem with the Kalam’s first premise. It is self-contradictory to propose that a causal principle itself ‘has’ to be caused; but allowing that it is not caused refutes the principle. Until the causal principle comes into existence, the principle that anything ‘has’ to be caused does not exist. So the causal principle itself cannot have had a cause, any more than there can have been a time before time, or a location north of the north pole. And as for that, so for anything else. One might try to argue that it is logically impossible for the causal principle not to hold, but Loke rejected that approach, admitting that it is not a logically necessary principle—wisely, as there is as yet, indeed, no argument that it is logically necessary. It is therefore contingent. It therefore needs a cause. Unless some things don’t need causes.

In fact, if that principle is contingent, then it is logically necessarily the case that there are logically possible conditions in which it will not exist and thus not apply; and therefore the first premise of the KCA is false. The most obvious such condition is the condition of nothing at all existing. But even the beginning of the principle itself, within any matrix of even something, is exempt from the principle, because the principle cannot precede itself. A contingent causal principle, therefore, cannot itself have had a cause. If it ever came into existence, logically necessarily, nothing caused it to. And if nothing can cause something (especially that thing), the requirement that ‘everything that begins to exist must have a cause’ is then falsified. Down goes the KCA.

Here I will first analyze the consequences of this; then analyze the debate, and how Loke specifically failed to escape these consequences. In a following article I explore a new way of conceiving of ‘nothing’ in the Argument from Nothing, one that ably intersects with science and gets to a far more plausible explanation of cosmogenesis than any that theism has ever proposed, which I came to from interacting with Loke and the other philosophers in discussion and Q&A, which was the most fruitful outcome of even having had this debate (and an audience of experts to engage with it). But here I will only summarize and conceptualize where the debate now stands.

Summary of My Position

What is notable is that most philosophers don’t challenge the first premise of the KCA; they usually expose the failure of one or more of its other premises. But I demonstrated the first premise is not in fact sustainable. It therefore should be rejected as well. This is not because the principle never operates (like now, over here, where we are located in spacetime); but rather, because it cannot have operated in the one set of circumstances the KCA requires it to (when there will have been no reality yet, and thus no spacetime, and thus no contingent laws or principles of physics at all, much less the causal). This is because the causal principle is not logically necessary; therefore, necessarily, it is contingent. And contingent things need a reason to exist other than themselves.

I showed that we can find such a reason in any logically possible nothing-state theists have to appeal to as pre-existing reality sans God: absent anything to dictate what will be, spacetimes governed by causal laws are statistically inevitable—something Loke even agreed with, declaring that if there were ever a condition not subject to the causal principle, then immediately “infinite” random universes would necessarily then arise (the actual point of my Argument from Nothing). I showed this was a logically necessary outcome of any logically possible nothing-state. Which means I do not have to reject the Principle of Sufficient Reason, since it is logically necessarily the case that a nothing-state will not be governed by a causal principle, and it is logically necessarily the case that in such a condition, an infinite multiverse will result. Which in turn provides a sufficient reason for nothing to become something—and in particular become, somewhere, a causally controlled something; and we can expect to always find ourselves there, as causally uncontrolled spaces can’t accumulate the complexity required to produce even stable life forms, much less self-conscious ones (see How the New Wong-Hazen Proposal Refutes Theism).

And So the Kalamity Collapses

So, in our debate, I demonstrated that the first premise of the Kalam Cosmological Argument is false, because it does not obtain in the one condition it needs it to obtain: when everything began to exist and no God was around at the time. In that one unique condition, there can be no prior causal principle to cause the causal principle; it therefore can only have arisen uncaused. And if it, so anything else. And as I showed, there is no way around this.

One could try to redefine ’cause’ so that any condition that produces the causal principle (like a nothing-state spontaneously generating it uncaused) counts as a cause; but then you have to abandon the fourth premise of the Kalam: that only a God can have caused it. When ‘nothing’ can have caused it, no God is needed anymore. And in fact ‘nothing’ is vastly simpler than a God and therefore, on Ockham’s Razor alone, is more likely to have caused it. This also makes many more successful predictions as to what we should observe now, in contradistinction to any pertinent God hypothesis.

One could instead try to insist a ‘nothing’ can never have existed—such as, perhaps, arguing (and this was suggested in Q&A) that nothing exists that does not exist somewhere at some time, and yet ‘nothing’ entails there is no place or time at which to exist. Ergo, there was never ‘nothing’. This is my own Argument from Nonlocality. Which also eliminates most gods proposed in cosmological arguments—because they, too, cannot preexist space and time (and thus ‘reality’ or even ‘the universe’), and thus by definition must ‘begin’ to exist. The conclusion then follows that, therefore, there can never have been ‘nothing’. But that logically entails there has ‘always’ been something, and once you go there, you must abandon the second premise of the Kalam: that everything began to exist. If it is logically necessarily the case that something always existed, then reality did not ‘begin’ to exist in any sense requiring a cause. It has just always existed.

This follows whether it is past eternal or not. For example, even if all reality began at some past time t, it has always existed in the same sense that it exists at every time there is (there is never a time when it didn’t exist; nor is there any place where it didn’t exist). This is the same sense in which theists need God to have ‘always existed’, since God cannot exist at a point before the first point of time. Therefore, for God to create time, he cannot precede it, but must coexist with it. But in that sense, the second premise fails to work for the Kalam, because if something necessarily always existed merely because it is impossible for there to have ever been nothing, then the thing that ‘always existed’ can be anything; there is no longer anything you need specifically a “God” to solve here. If something just always existed, it can be literally anything. This is how theists exempt God from the second premise: even though he ‘begins’ to exist in the sense that he did not exist before the first point of time (because that is logically impossible), they insist he nevertheless in some sense ‘always’ existed. But if that is true for God, it’s true for the universe itself, without God. The second premise of the Kalam therefore is either false, or includes God; and any method of getting around God then falling afoul of the first premise will then work for anything else that isn’t God.

This is why cosmological arguments tend to covertly depend on ontological arguments, and therefore are always circular arguments: they require the premise that God necessarily exists (and thus ‘exists’) simply to get the conclusion that he exists; but then the conclusion is in the premise. Otherwise, God is a contingent being, and so also requires a cause like any other contingent being. So God has to necessarily exist. But if you could actually prove that premise (that God necessarily exists), you have no need anymore of the cosmological argument. It does no work. The same fate befalls even fine tuning arguments (see A Hidden Fallacy in the Fine Tuning Argument).

For all these reasons, Kalam is a mobius loop of whack-a-mole apologetics: each premise will be defended in isolation using tactics that negate the other premises; but because each premise is being defended separately, no one notices this (until, of course, they do). Theists will thus, when cornered on one premise, jump to another and ‘forget’ they were cornered on the other premise. Any argument with at least three premises thus allows a delusional death-loop, whereby one jumps to another premise as soon as any premise is fatally challenged; and because they are always at least two steps of memory away from what started this, they can keep clear from their memory that their defense of a premise two steps of reasoning ago entailed the refutation of the step they have retreated to defending instead. 

It thus becomes easy to delusionally go on believing the Kalam argument works, when it never does. It is fundamentally incoherent. There is no definition of its terms on which all of its premises are true.

Summary of Loke’s Position

Loke’s central premise (as in his book) was (with my explication in brackets):

• If x (e.g., physical reality) begins uncaused, then [any other] y which [could] begin to exist would also begin uncaused.

Then, since we don’t observe this now (there are no spontaneously-starting y-things), then it can’t have been the case at any prior time either (by modus tollens). But this premise is false. It is not the case that ‘if time began uncaused, therefore we should be observing all sorts of things in time beginning uncaused’ because the conditions are not analogous. That there will be no temporal phenomena (like cause-effect principles) when there is no time is necessarily true, yet does not entail or even make probable that there will be no no temporal phenomena when and where there is time. To the contrary, when and where there is time is precisely where we should expect to observe temporal phenomena like causal principles; whereas when and where there is no time is precisely where we should expect not to observe temporal phenomena like causal principles.

This fully generalizes: that there will be no principles (like Loke’s principle of causation) when and where there are no principles does not allow the inference that there will no principles (like Loke’s principle of causation) when and where there are principles. So we should expect to observe a y when there is nothing to stop it; but we should not expect to observe a y when there is something to stop it—as there is: ordered spacetime now exists and thus limits what can appear or disappear or in any other way change. It is therefore not the case that ‘if x, then y‘, because x refers to a state of affairs that does not exist at y, and yet it is that very state of affairs that produced x; the subsequent absence of that state of affairs then prevents y.

Loke could never get around this. He attempted some semantic tricks that bombed—like trying to argue a ‘law of physics’ is not a ‘principle of physics’ in order to avoid the fact that no laws exist; that’s a non sequitur, because even if you could change what a thing was by changing what you call it (and you can’t; semantics has no effect on ontology), it is still the case that when there is nothing there are no principles either. So even trying to define a principle into being something different from a law, he didn’t escape the consequence of the fact that a principle is either logically necessary or contingent; he admitted this one was contingent; and contingent principles cannot ‘always’ exist. That’s the whole point and consequence of not being logically necessary: that entails there are conditions in which the thing will not exist. And one obvious place principles won’t exist is when nothing at all exists; another, is any time when those principles do not yet exist; yet another, is when the ontological entities necessary to cause that specific thing to exist, don’t exist. And these being the case eliminate the first premise: it simply is not ‘always’ the case that everything that begins to exist must have a cause; and the conditions in which it won’t be true are precisely those that the Kalam is trying to prove with this premise don’t exist.

There is no escape from this loop. The dog can chase its tail forever; but it will never catch it.

Stymied, Loke had only one other recourse: to attempt to argue that my ‘nothing-caused’ model entails predictions that aren’t born out, and therefore it must be empirically false—and therefore, his central premise must be true. This approach doesn’t even logically work, because my argument was that his premise being true is logically impossible. It therefore cannot be refuted empirically. Bringing an inductive argument to refute a deductive proof is like bringing a sponge to a gunfight. It is simply logically impossible that a causal principle will exist when nothing, literally nothing, exists (other than what is logically necessary and thus not subtractable even from a total state of nothing, a point I have emphasized in all my prior discussions of the Argument from Nothing, and which Loke continued to ignore; but I’ll have more to say on that point in my next article, which deals with a positive demonstration of a possible naturalist cosmology rather than, as here, a negative refutation of a bad argument for God).

Loke’s approach was thus another instantiation of that same circular argumentation: he needed it to be necessarily the case that his principle was true (and therefore the Kalam’s first premise a logically necessary truth), but he admitted he could produce no case for that; yet his argument, that the absence of a causal principle before reality comes to exist would predict observations we aren’t seeing now, was, in effect, arguing that it was logically impossible for there to have ever not been that principle. But he was doing this inductively, which is impossible. If it is not logically necessary, then it cannot be logically necessary that the observations he claimed we should be making would be observed; and yet if it is not logically necessary that the observations he claimed we should be making would be observed, then there is no reason to expect to observe them in the first place. Loke was thus, again, arguing in a circle.

Either the predictions he claimed would follow are logically necessary, or we might not see them; but if they are logically necessary, so is his principle, but he could not show that it was; whereas if we might not see them, then our not seeing them is no longer evidence that the principle is logically necessary (a.k.a. ‘always true’). Hence Loke at more than one point confused this (the first cause argument) with the fine tuning argument. He wanted it to be probable (not necessary) that we would observe the things he claimed unless his causal principle was necessary and not contingent; but that cannot get the result that the causal principle was necessary and not contingent.

Loke would thus slip into arguing things like, in effect, ‘yes, in some cases we would not see these things I claim we would see, but it’s very improbable we would find ourselves in one of those cases; therefore, our location must have been finely tuned ‘just so’ as not to see what we expect; therefore the causal principle always holds’. Notice the non sequitur: the conclusion ‘therefore the causal principle always holds’ does not follow from an admission of our being at an improbable location; all that follows from that is that we are at an improbable location. That does not argue anything at all in favor of the causal principle ‘always’ being true, as the first premise of Kalam requires. Loke was thus confusing inductive with deductive argumentation, and ran in circles this way across the whole debate. Yes, maybe there is something weirdly coincidental about where we ended up. But…so? We weren’t debating that. And regardless, that tells us nothing about whether the causal principle ‘always holds’, the thing we were debating.

Loke’s Four Predictions

So Loke’s entire mode in the debate was illogical. He literally couldn’t win that debate even in principle. By confusing inductive for deductive arguments, and mistakenly thinking you can rebut a deductive proof with an inductive one, he never succeeded in even answering the point that principles cannot exist when nothing exists—and therefore his principle cannot have preceded reality so as to have applied to its own origin. The First Premise of the Kalam is simply logically false.

Nevertheless, even his irrelevant inductive arguments were false. Indeed, they were simply repeated from his book, as also in his prior published reply—so he produced nothing new in that reply or this debate; I already refuted all this in my original article. But let’s go over them again: he claimed ‘four’ predictions of my model; but none of them were predictions of my model. It actually predicts the opposite of them.

• (1) We should still see spontaneous creation. False.

Once ordered spacetime exists, what can or can’t materialize within it is limited by the properties of that spacetime. This is not the case when there is no spacetime at all, much less an ordered one. There is no reason to expect “spontaneous creation” to be a property of an ordered spacetime—beyond what we already, indeed, observe to be the case (see All the Laws of Thermodynamics Are Inevitable), which illustrates the problem: we do see y, all the time; only, it is limited to a small set of ultra-simple subatomic particles, constrained by wavelength. In spacetime, only spontaneous wavelengths can materialize and vanish at random; every possible subatomic particle is covered by one or another wavelength and thus simply is what results when such a wavelength is realized in our spacetime; and an aggregate average of this is what we in fact observe, as the virtual particle background (as statistically a random collage of wavelengths and antiparticles cancel each other out, especially at scale, and we live at a vast scale, some 10^30 to 10^40 times above the smallest units of spacetime: indeed, that is probably why organized life only appears at such a vast scale: the averaging to a dim baseline is pretty complete by then, and thus laws of causation and conservation emerge at this scale, and not that one).

The reason we don’t see just any other things here (other particles, other entities) is that ‘here’ is a specific spacetime manifold that only allows certain spacetime-specific phenomena to appear. You thus can’t have a ‘Jedi Force particle’ because there is nowhere here for one to exist; spacetime lacks any structure requisite to produce it. Causal laws emerge from the laws of thermodynamics, which simply describe the inevitable outcome of a large-scale randomization of what happens—in other words causal laws emerge from the absence of causal laws, once enough random ‘things’ exist to average out most and allow the survival of only the lucky few. So, we actually do see what Loke predicts (virtual particles constitute his y); but we don’t see what Loke predicts in any other sense. And the reason is that an ordered spacetime now limits what can happen—all laws and principles of physics derive therefrom. That’s why, contra Loke, the strength of the electromagnetic field can’t just “spontaneously” change: it is a property of spacetime itself (literally, the permittivity of the vacuum); so it now needs a cause. But when the thing causing it didn’t exist, it wouldn’t have needed a cause.

I illustrated this in the debate with this Big Bird slide:

Obviously, when nothing exists, neither will causal or conservation laws; but when something exists, then and there we predict there will be causal and conservation laws. So my model predicts the opposite of what Loke avers: it predicts that if then x, then now not y. So his first inductive argument was simply false: my model does not predict what he claims. It predicts the opposite. It predicts exactly what we observe.

• (2) We should still see spontaneous violation of conservation laws. False.

Conservation is just another reference to causation. To say something is conserved is to say nothing happens uncaused, and therefore the conserved quantity or quality will not change in the absence of a suitable cause of it. So this argument falls to the exact same objection as just articulated. An ordered spacetime now exists to control conservation, just as it now controls causation. But when that ordered spacetime didn’t exist, it didn’t. So we expect the opposite of what Loke avers: if x (reality) happened uncaused (because then nothing existed to manifest any causal or conservation laws), then y (anything else) will be governed by some causal and conservation laws (because now something exists to manifest them).

As I pointed out last time, when discussing why Stephen Hawking didn’t “really” describe a universe-from-nothing cosmological model, Loke himself argues (TKCAR p. 259) that Hawking “failed to mention that at the subatomic level quantum particles do not come into existence from absolutely nothing; rather,” they “are manifestations of pre-existent quantum fields which act according to pre-existent quantum laws.” Bingo. Loke already knows his own core premise is false: it’s when x, not y—because when x, there were no “pre-existent quantum fields which act according to pre-existent quantum laws,” but now that there are, they constrain what can and cannot happen, thus preventing any y (other than the very y they allow, and which indeed we observe, confirming this fact).

Spacetime is not empty. It is full of ontological entities, forces, and qualities. Yet what can or can’t happen or exist is limited to being able to happen or exist in spacetime—and not just any spacetime, but our particular spacetime, those specific “pre-existent quantum fields which act according to pre-existent quantum laws.” Loke thus intuitively understands that a true nothing will not be governed by those laws, or thus any laws. He thus intuitively knows his core premise is false. He just can’t admit it, causing him to twist himself up into irrational (and as I noted last time, often dishonest) circles.

• (3) We should observe inductive reasoning to routinely fail. False.

As I noted, life will only arise to observe itself in a world sufficiently ordered to generate life. But any ordered world will be predictable to some degree. Induction is the logic of prediction. Therefore, induction will work in all worlds in which life arises to observe itself. This is the actual prediction of my model. Loke’s prediction is thus false—we expect the opposite (see The Argument from Uniformities). Loke tried to get around this with an inept argument from inverted Boltzmann worlds, revealing that both he and his cited scholar are bad at math and science. Which is probably why they are still Christians.

To understand what an ‘inverted Boltzmann world’ is, you have to understand what a Boltzmann world is. I’ve discussed this extensively elsewhere (see The Boltzmann Brain Argument), but in short, a Boltzmann world is a world—let’s say, one identical to ours, as you observe it right now—that has, by mere chance accident, come together as if there has been a whole past causal history to it, when there wasn’t; it didn’t exist half an hour ago. I’s just that, by random chance, it collided into existence in just luckily, precisely a way that looks like it didn’t. An inverted Boltzmann world is identical to that one, only backwards: it spontaneously, accidentally, just falls apart in half an hour and thus ceases to exist, despite our predicting that it would keep going.

There isn’t any logical basis for inverted Boltzmann worlds to undermine induction—induction keeps working until it doesn’t, so there isn’t any reason to give up on it until it doesn’t matter anymore. So the only way to get this to be a problem is, first, to insist that an inverted Boltzmann world does not just accidentally dissolve in half an hour (which, self-evidently, requires a conjunction of chance accidents with a statistical probability of functionally zero, and so induction correctly predicts that won’t happen to any probability worth caring about), but was never really held together at all and so it will inevitably dissolve in half an hour (all the atomic bonds, gravitational equilibriums, and everything else are themselves not even real, but only apparent by accident); and, second, to insist that such worlds are likely, indeed more or as likely as causally stable worlds (where atomic bonds, gravitational equilibriums, and so on are all real—so even if they formed all at once by accident, they will go on working as predicted thereafter). But this conjunction of claims will always be illogical.

The number of chance accidents required to produce a causally stable world from a random beginning (like a Big Bang) is vastly (and I mean vastly) lower than the number of chance accidents required even to produce a causally stable Boltzmann world, and that is vastly (and I mean vastly) lower than the number of chance accidents required to produce an inverse Boltzmann world that began causally stable—and that is vastly (and I mean vastly) lower than the number of chance accidents required to produce an inverse Boltzmann world that was never causally stable (the kind required for Loke’s argument, where all the features holding the world to a causal course don’t exist but merely appear to, by an extraordinarily convoluted number of yet additional chance accidents). So in an infinite randomized array of universes (in which all these kinds of worlds will inevitably exist), the probability we will find ourselves in an unstable inverse Boltzmann world is so small as to be functionally zero. It therefore poses no threat to inductive reasoning.

Loke tried to rebut this by appealing to the measure problem, but that reveals he does not understand the mathematics. The measure problem, simply put, is the problem that we cannot calculate an objective probability of something when dividing by infinity—which is why it is usually solved by using finite temporal slices of a multiverse’s growth, although any finite random sampling of the possibility space would work as well, to a measurable statistical certainty. Another aspect to the measure problem is the problem of knowing what to count; but that doesn’t affect permutation-spaces where the same things are being counted, as in this case. The fact that Boltzmann worlds will be vastly less common than causally stable worlds requires no division by infinity, and faces no question of what comparatively to count, and is therefore immune to the measure problem. The fact simply is that a causally stable world contains (requires) fewer Feynman actions in the path integral describing it than a Boltzmann world. A Boltzmann world requires vastly more coincidental events (e.g. all the random relocations or appearances or disappearances of subatomic particles in just the right way) than a causally stable world does.

For example, the number of random accidents needed to causally produce a rabbit in the usual way is vastly smaller than the number of such steps to just instantly, at random, ‘produce a rabbit’. For example, all the DNA in all its cells is already there in the first case; it doesn’t all have to be randomly assembled from scratch. In fact almost all of the steps in the first case have near 100% probabilities (thanks to causal determinism); whereas all those same steps have astronomically low probabilities in the second case (thanks to the absence of causal determinism: when you have to get an outcome by chance accident rather than a deterministic causal pathway), and the laws of probability necessarily entail the cascading difference in total probability is going to be absurd.

Indeed, as I explained last time, a Big Bang has a probability of random assembly of only 1 in 10^10^10^56, whereas even just one Boltzmann rabbit has a probability of random assembly of far less than 1 in 4^2,000,000,000^20,000,000,000,000. To be precise, I ran these numbers for a tiger, and only for its random DNA assembly, not all the additional assembly of cells and atoms and subatomic particles; so the probability even for a tiger is far lower, and a rabbit is at this resolution comparable in scale to a tiger. But this means a Big Bang will randomly happen in our universe over 10^1,000,000,000,000 times more often than even a single Boltzmann rabbit popping into existence—much less an entire Boltzmann world (which requires a lot more accidents than a mere single rabbit).

So we don’t need to determine what “the probability” is of a Boltzmann world, or any other; which is good, because the measure problem makes that impossible in a suitably randomized infinite multiverse. Rather, we only need to know what the relative probability will be of a Boltzmann world compared to a causally stable one. And we can do that with simple permutation theory: just count all the chance accidents that are required in the one case that are not required in the other. Both numbers are finite; therefore there is no division by infinity. And the same things are being counted (all else is equal). The difference in count will then correlate to the difference in frequency of those outcomes encountered when relying on random selection alone.

So, there may be infinitely many Boltzmann worlds, and infinitely many causally stable worlds, making identifying their total ratio impossible; but any finite random sampling of that possibility space will produce a vastly larger number of causally stable worlds than of Boltzmann worlds, to an extremely high probability (and thus confidence level), because any count of possible worlds will contain vastly more of the one than the other. Another way of putting it is that Boltzmann worlds have vastly higher specified complexities, and thus of all random configurations, they will count few; whereas causally stable worlds can be formed in vastly larger numbers of ways, and thus of all random configurations, they will count many. This is why the Second Law of Thermodynamics always emerges from any randomized system; it is literally the same concept: highly ordered results are fewer in number among all possibilities than relatively less ordered results—like structurally randomized Big Bangs. As low as the Big Bang’s entropy was, that entropy count was almost solely in the ordered location of its energy (squeezing so much into a single small space), not its organization. If it had to also be precisely organized, its entropy would also have to have been vastly (and I mean vastly) lower still. You can get the randomized low entropy state roughly 1 out of every 10^10^10^56 times of trying; but to get an even more specifically ordered result than that (like a sudden accidental rabbit), you’re looking at over 10^1,000,000,000,000 more times of trying before you get one. Therefore there will be over 10^1,000,000,000,000 times more causally-ordered “Big Bang worlds” than any kind of Boltzmann worlds (inverted or otherwise).

In other words, there are vastly many different ways to arrange the energy in the Big Bang’s initial state that will have functionally the same result (e.g., think of all the ways galaxies could have been formed and arranged in respect to random perturbations in the inflation event that would be functionally the same universe as this). To use the same analogy as last time: a rabbit soup has low entropy (we’ve assembled a complex of rabbit parts and other ingredients into a small space, like a bowl, and heated it); but it has vastly higher entropy than a living rabbit, where the organization of parts is extraordinarily more specific. There are a gazillion more ways to mix up a rabbit soup and still have rabbit soup than there are ways to mix things up and still have a living rabbit. There are just fewer configurations that get you that, than get you rabbit soup. Vastly fewer. Which is why rabbit soups aren’t alive anymore.

So an infinite multiverse poses no problem for us. Induction can be expected to work here, to a probability so near to 1 as to make all odds. In fact, since all Boltzmann worlds are just another kind of Cartesian Demon, everything I have said proving all Cartesian Demons too improbable to pose a threat to inductive reasoning (in We Are Probably Not in a Simulation) applies to Loke’s desperate clutching at the vastly improbable to try and ironically get a result to be probable. That doesn’t work. Vastly improbable outcomes are simply vastly improbable. And Loke presented no evidence to change that estimate of probability, much less get it anywhere high enough to “be a problem.”

• (4) We should observe far more collisions with other universes than we do. False.

Loke never made any scientific sense out of this argument. He seems not to understand the science at all. Almost all going cosmological theories now entail infinite multiverses. Yet Loke could produce no peer reviewed cosmological finding, anywhere, of a prediction of a higher collision count than we observe. He thus could never justify how or why this should result from an infinite multiverse. His argument was both antiscientific, and wholly contrary to fact, and illogical.

In most conceivable multiverses, there is no meta-extra-space in between them through which they move and thus in which they can collide at all. Without a universe in which to occur, there is nowhere for a collision to occur. To get universes to collide, you have to posit an additional epicycle of an extra-meta-space in which universes move and thus can interact, and you have to posit this over and over again, millions and billions of times, to get your extra-meta-space so full of potentially colliding universes that you get a high expected count of collisions. The probability against this is astronomical. Chosen at random, most multiverse regions will not have this massively convoluted extra space between them.

Loke also couldn’t produce any evidence that if there were so many, that it would not match what we already observe. For example, see Ambjørn & Watabiki, “Is the Present Acceleration of the Universe Caused by Merging with Other Universes?” Journal of Cosmology and Astroparticle Physics (2023), and the summaries “Multiverse Collisions May Dot the Sky” at Quanta and “Will Our Universe Collide With a Neighboring One?” at Discover. But it is also not expected to be the case. For example, in Many Worlds Theory, all universes are moving away from each other, not colliding; and in Eternal Inflation Theory, most inflation events are also moving away from each other, whereas internal spontaneous inflations will be so rare we can expect never to see one. And, of course, we also won’t likely be where multiverse collisioning is a problem—as then we wouldn’t be here. Natural selection will have left us somewhere else, in a pocket of the multiverse more stable for life to form.

So, again, my model made no such prediction as Loke claims. It makes, actually, the opposite prediction. And indeed, even if the prediction were somehow statistically true (i.e. if we could show—though no one has—that multiverse collisioning is more frequent than not and thus should be expected), then that prediction has been fulfilled, contrary to Loke’s insistence that it has not: per the above studies and analyses, we might even be in a stable collidingverse, the only kind life can expect to find itself in. Which would then confirm an infinite multiverse. And that would definitely do Loke’s God in. More importantly, even if we could show—though no one has—that the multiverse collisioning we observe is much less than expected, this would have no effect on the logically necessary fact that the causal principle cannot have existed before reality (and thus principles governing a reality) existed. We’d then just be in a rare location in the ensuing multiverse. Which fact, however weird, would have no bearing on the former logical necessity.

Conclusion

Loke couldn’t demonstrate his required premise to be true. Whereas I proved it to be logically impossible. Consequently, Loke failed to rescue the first premise of the Kalam Cosmological Arrgument. That premise is false. There simply can be conditions under which the causal principle doesn’t govern. And the origination of all reality, and hence the causal principle itself, is one of those conditions, since a causal principle cannot have preexisted itself so as to be required to bring it into existence. Therefore, being contingent, the causal principle must have had a beginning (or could only have always existed by chance accident, not necessity); and therefore must itself have been caused (or could itself have once not existed); but that is impossible when the principle doesn’t yet exist. The first premise of Kalam is therefore false, and false exactly when Kalam needs it to be true to get to God. Loke tried to avoid admitting this by recourse to (1) bogus semantic maneuvers, (2) an illogical confusion between inductive and deductive argumentation, and (3) false claims as to what the alternative predicts.

-:-

Loke has since continued to reply by repeating over and over again the points I already refuted, and ignoring everything I actually said in refutation of them. So I see no value in any further engagement with him. On my last discussion and why I’m done with him, see my video with Godless Engineer.