Why isn’t Atheism the default position according to William Lane Craig?

I found this interesting passage by Craig in A Debate on God and Morality, which is about a debate that Craig had with Wielenberg:

Platonism is a metaphysical view which is so extravagant that it makes theism—which itself involves hefty metaphysical commitments!—look modest by comparison …. Given this strange bifurcation of reality into these two causally unconnected domains, it would be much more credible to suppose that one of the categories is empty. But concrete objects are indisputably real and well-understood, in contrast to abstract objects. So, van Inwagen maintains, the presumption should be that abstract objects do not exist. Nominalism of some sort is thus the default position. (Craig, 2020)

This got me thinking: Shouldn’t atheism be the default position or presumption for the same reason? After all, Craig does acknowledge that theism “involves hefty metaphysical commitment.” It’s just that it’s modest in comparison to Platonism. Yet elsewhere Craig writes:

Certain atheists in the mid-twentieth century were promoting the so-called “presumption of atheism.” At face value, this would appear to be the claim that in the absence of evidence for the existence of God, we should presume that God does not exist. Atheism is a sort of default position, and the theist bears a special burden of proof with regard to his belief that God exists.

So understood, such an alleged presumption is clearly mistaken.  For the assertion that “There is no God” is just as much a claim to knowledge as is the assertion that “There is a God.”  Therefore, the former assertion requires justification just as the latter does.  It is the agnostic who makes no knowledge claim at all with respect to God’s existence.  He confesses that he doesn’t know whether there is a God or whether there is no God.[i]


[T]here is a huge difference between an atheist and an agnostic. The agnostic makes no knowledge claim at all. But the atheist is making a knowledge claim. He is saying God does not exist. That is a knowledge claim that requires an argument or justification in the same way that the claim God does exist. [ii]

Two points. First, Craig makes a point that the atheist appears to be using some kind of faulty absence of evidence reasoning. I’ll grant that, as it’s not important for my point. My question is: Why can’t the atheist reason just as Craig reasoned with respect to Platonism? That is, the theist is making a “hefty” metaphysical claim that is far different from the other things we think exist, so the default position is that God does not exist.

Second, Craig says that the atheist and theist are making knowledge claims so it requires an argument or justification. This suggests to me that he thinks agnosticism is the default position, and that it doesn’t require argument or justification. If I’m reading this right, then it seems Craig should think agnosticism is the default position with respect to abstract objects.


Craig, William Lane, and Erik J. Wielenberg. A Debate on God and Morality: What is the Best Account of Objective Moral Values and Duties?. Routledge, 2020.

[i] https://www.reasonablefaith.org/writings/question-answer/definition-of-atheism/

[ii] https://www.reasonablefaith.org/podcasts/defenders-podcast-series-3/doctrine-of-god-part-13/excursus-on-natural-theology-part-28/

Posted in Philosophy of Religion | Tagged | Leave a comment

William Lane Craig’s moral argument and the speed of light

There’s this fairly well-known moral argument for God that’s popularly defended by William Lane Craig. In this short post, I want to present what I think is a novel rebuttal. Let’s look at Craig’s moral argument:

  1. If God does not exist, objective moral values and duties do not exist.
  2. Objective moral values and duties do exist
  3. Therefore, God exists.

I want to focus on the first premise: If God does not exist, objective moral values do not exist. Now let’s suppose God doesn’t exist, but a demigod does. Demigod is a necessary being, has the same moral virtues that God has, is equally omniscient, and is very powerful, yet he is not quite omnipotent. For example, let’s suppose demigod is not omnipotent because he cannot make things go faster than the speed of light. I think there are two main conceptions of the philosopher’s God: (1) the greatest conceivable being, or (2) an omnipotent, omniscient, morally perfect and necessary being. Demigod seems to not satisfy either of these conceptions. So, Craig has to say that if God did not exist, and demigod did exist, then objective moral values and duties would not exist.

This seems to be an odd consequence of Craig’s 1st premise. God and demigod are very much alike, save for the fact that demigod can’t make things go faster than the speed of light. How can this difference be a difference-maker when it comes to objective moral values and duties? What do facts about making things travel faster than the speed of light have to do with objective moral values and duties? Since demigod can’t make things go faster than the speed of light I don’t have an obligation to save the drowning baby? The only relevance power might have is the need to hold people morally accountable in terms of heaven and hell, yet demigod has sufficient power for that.

The goal here is to point out that Craig is making a fairly bold claim in this deductive argument. He’s not merely saying that God would better explain or ground morality than demigod or other views would; he’s saying that only God could do that, and it’s far from obvious why demigod can’t do just as well in grounding morality.

Posted in Metaethics, Philosophy of Religion | Tagged , | 2 Comments

Some remarks on Josh Rasmussen’s How Reason Can Lead to God

I recently read Josh Rasmussen’s recent book How Reason Can Lead to God. The book is accessible for the non-philosopher, but that’s not to say that the book doesn’t touch on many deep philosophical issues. Rasmussen says, “arguments do the most good when they are tools that promote an inquiry .” I think that’s exactly right, and Rasmussen’s arguments certainly do that. Rasmussen and Leon’s sister book Is God the Best Explanation of Things has a lot of the ideas of the first book and it goes much deeper. It’s set up as a dialogue between two expert philosophers—a theist and an agnostic. This one is more technical, and is aimed more towards those with some philosophical background; it helped me see some of the issues from the first book more clearly.

Rasmussen’s main point behind the two books is that we can use reason to discover that there must be a perfect foundation to reality. Reason further provides us with its many of its attributes. He summarizes his argument like so:

Premise 1. Reality in total is self-sufficient (with no outside cause or explanation).
Premise 2. Nothing can be self-sufficient without a perfect foundation.
Conclusion. Therefore, reality has a perfect foundation.

The idea behind premise 1 is fairly simple: nothing exists outside of reality (if it existed it would be in reality), therefore nothing outside reality can cause or explain reality, which is what it means to be self-sufficient. Premise 2 is supported in two ways: (1) the problem of arbitrary limits, and (2) the problem of construction.

The problem of construction is about looking at certain things in the world—minds, matter, morals, reasoning—and seeing that some foundation hypotheses lack the materials to construct those things. For example, he argues that just as white tiles are the wrong materials to construct a purple floor, molecules are the wrong materials to construct a mind, and likewise with matter, morals, and reasoning. The problem of construction is a big topic, so I intend to focus this post on the aspect that peaked my interest: (1) the problem of arbitrary limits, and corresponding issues with simplicity and intrinsic probability.

Rasmussen thinks that the foundation of reality cannot have unexplained arbitrary limits. First, what is an arbitrary limit? Rasmussen explains:

What I want to say, more precisely, is that a limit is arbitrary in this sense: The limit could conceivably have been slightly greater or slightly lesser. For example, if the foundation was the shape of an octagon, then its shape is arbitrary because it could conceivably have had nine vertices instead of eight. Why eight? That number is arbitrary in view of the conceivable alternatives (Rasmussen & Leon, 2019, p. 113-114).

On the view that a “limit could conceivably have been slightly greater or slightly lesser,” it seems even a perfect foundation is arbitrarily limited, since I could conceive of it being slightly less perfect. This is not a welcome result for Rasmussen.  Suppose, instead, that we say something is an arbitrary limit if it could conceivably have been greater. This would fix the problem for a perfect foundation because nothing can be greater than perfect, but there is still another issue. Leon points out that a problem for quantitative features is that there are always greater infinities such that every quantitative feature would be limited. So I suggest we could understand Rasmussen’s view that an arbitrary limit is something that is not maximal in its fundamental qualitative features, as the following passage suggests:

My proposal so far, then, is that a maximal foundation would have no fundamental feature along a continuum of potentially (i.e., conceivably) surpassed magnitudes.
I should also clarify that maximal features are not the same as infinite magnitudes. I agree with Leon that even infinite magnitudes are conceivably surpassed. The fundamental features, then, would be nonquantitative features, like being independent, being foundational, having supreme power, and so on. (Note: having supreme power is not a quantity of power, but a quality or kind of power; or, if we want to call it a “quantity,” it a special, unsurpassable quantity (Rasmussen & Leon, 2019, p. 140-141).)

Given that properties like mass, number of vertices, size etc. are quantitative, they will count as limit; i.e. even an infinite mass, space, or velocity will be limited. My interpretation, then, is that anything that is not qualitatively maximal is a limit and hence requires an explanation.[1] Rasmussen summarizes his argument:

All limits alike have an outside explanation upon which they depend. Yet the basic features of the foundation, by contrast, lack an outside explanation. Therefore, the basic features of the foundation have no limits (Rasmussen, 2019, p. 72).

This still leaves open that the foundation has limits that are explained. What Rasmussen objects to is unexplained limits. For example, if the foundation was the Trinity, the quantity of 3 persons could be explained by (perhaps) Swinburne’s suggestion of highest love.

Rasmussen thinks there are three reasons to think the foundation lacks arbitrary limits: (1) limits are less simple and hence less intrinsically probable, (2) limits block explanatory depth since that limit itself won’t be explained, and (3) limits are uniformly dependent such that they all require an explanation. I’ll go through each in turn.


Earlier in the book Rasmussen writes:

The complexity of a hypothesis flows from the quantity of basic components within it, not the quantity of words used to express it. Consider, for example, the following hypotheses: (1) “Adam is a bachelor” and (2) “Adam is a male.” Both hypotheses have four words. Still, the word bachelor contains more information than the word male. In fact, the concept of a bachelor includes the concept of a male plus the concept of being unmarried. For this reason, hypothesis (1) is more complicated than hypothesis (2). Simplicity, then, is not about how many words you use. Rather, it is about how many basic concepts are involved. The fewer the basic concepts, the simpler the proposition (Rasmussen, 2019, p. 63).

I think ‘perfect’ is conceptually or semantically simple—like Moore’s ‘good’, ‘perfect’ seems semantically irreducible—but I don’t think a perfect foundation hypothesis is as simple as it may seem. A lot of preconditions need to be in place for a being to be perfect, which lowers the intrinsic probability. For example, a perfect being must be at least omnipotent, omniscient, and morally perfect—in addition to other perfect-making properties; and, in order to be morally perfect, the being must have many good-making properties like being loving, generous, honest etc. To illustrate the hidden complexity, let’s compare two hypotheses:

H1: The foundation is perfect.
H2: The foundation is omnipotent and omniscient.

There’s a sense in which H1 is simpler—it has fewer basic concepts—but H2 is more intrinsically probable. Compare:

H1’: Bob knows that it will rain tomorrow.
H2’: Bob has a justified belief that it will rain tomorrow.

H1’ is simpler than H2’ in that it has less basic concepts, but H2’ is more intrinsically probable than H1’ since having knowledge entails justified belief but not vice versa. Concerning intrinsic probabilities, P(agent has a belief that p) > P(agent has a justified belief that p) > P(agent has a justified true belief that p). Having a justified true belief is a precondition to knowledge, and, similarly, the perfect-making properties are preconditions to perfection.

‘Making’ should be understood in a metaphysical grounding or explanatory sense rather than a causal sense; the perfect-making properties metaphysically ground or explain perfection. This can be shown with a Euthyphro-like question: Is the foundation perfect because it is omnipotent, omniscient, etc. or are these properties perfect-making because the perfect foundation has them.[2] I think this direction of explanation is important because this means the perfect foundation hypothesis doesn’t explain the perfect-making properties like independence (i.e. not depending on something other); rather, the foundation is perfect partly because it is independent.[3] Each additional perfect-making property adds to the complexity and hence lowers the intrinsic probability of the perfect foundation hypothesis.

Rasmussen explains that simpler hypotheses include less information and so have fewer ways to go wrong:

[The] reason the simpler hypothesis is more internally probable—is that the simpler hypothesis includes less total information and so has fewer ways to go wrong. To illustrate, suppose you have two hypotheses A and B. Suppose, next, that A is composed of basic parts, a1 and a2 and a3, while B is composed of just b1 and b2. Then A has three “opportunities” to be false, while B has only two. Now suppose you have no idea whether any of the components of A or B are true, and you have no idea whether the truth of one component depends on the truth of any other. Each component is equally plausible in your mind. Then, from where you stand, B is more probable than A because B has fewer components—and so fewer opportunities to be false (Rasmussen, 2019, p. 62).

Each perfect-making property—a precondition for perfection—adds information such that there are more ways things can go wrong. It’s not easy being perfect.

If I’m right, the simplicity of the perfect foundation hypothesis will depend on the simplicity of the cluster of perfect-making properties. Presumably Rasmussen will think that—following the arbitrary limits principle—power, knowledge, and moral value will be simpler if they are unlimited (i.e. maximal) rather than limited. This would be a favorable result for the perfect foundation hypothesis. In similar fashion, Swinburne thinks the God hypothesis is simple, based on what Gwiazda calls Principle P:

Hypotheses attributing infinite values of properties to objects are simpler than ones attributing large finite values. (Gwiazda, 2009)

Others have offered reasons for thinking the omni-properties are simple as well (Draper 2016; Miller 2016). Simplicity is being used as a guide to the intrinsic probability of the hypothesis: the simpler the hypothesis the higher the intrinsic probability.

An alternative way to assign intrinsic probabilities is to use the principle of indifference, where we apply a uniform probability distribution among the possibilities. For example, Tooley (2009) argues that, given an omnipotent and omniscient being, we should assign a 1/3 probability to the moral properties of being good, indifferent, and evil, such that the intrinsic probability of theism is less than 1/3. Similarly, we could apply the principle of indifference to power, cognitive ability etc. The result is that the intrinsic probability of the omni-properties is the same as any other particular degree of that property. This means that the intrinsic probability of the omni-properties is highly improbable, since there are many more ways to be limited than unlimited.

I don’t wish to settle the best way to assign intrinsic probabilities. I merely wish to point out that the principle of indifference is a plausible alternative method that would further lower the intrinsic probability of the perfect foundation hypothesis.

Explanatory depth

Consider next the problem of explanatory depth, which is another problem for an arbitrarily limited foundation.

The problem here is that if the foundation has arbitrary limits, then the foundation (or a theory of the foundation) has less power to explain the things that depend on it. Recall the theory just given: the foundation has (1) a mass of 241 grams, (2) the shape of a sphere, and (3) the capacity to produce exactly 288 particles. Focus on its mass. What might explain the fact that the foundation has a mass of exactly 241 grams (Rasmussen, 2019, p. 69)?

The point here is that if the foundation has 241 grams, not only is that left unexplained, the existence of mass in general is left unexplained.  I’ll only offer a tiny sketch of a reply here, as I think it would take a lot of work to go through all Rasmussen’s points. I am the least confident in what I have to say here.

If God existed, I suppose the explanation would go something like this: God desires the good, and conjoined with his beliefs and powers, this would explain the good things we see. Even if the perfect foundation hypothesis is the best and only explanation for why there are limited things like shapes and mass, I feel like the perfect foundation hypothesis is so vague that the explanation isn’t good enough to warrant adding the perfect foundation onto our ontology.  I do not clearly see that the perfect foundation predicts shapes and mass as a means to the good. Every theory of the foundation will have some brute facts, and we can try to explain those brute facts, but unless that theory has enough theoretical virtues, we shouldn’t add it onto our ontology.


Rasmussen’s final tool against unexplained arbitrary limits at the foundation is uniformity:

To illustrate the problem exposed by uniformity, imagine a mountain range. This mountain range has a particular shape along its mountaintops. No matter the shape, it has some explanation …. Whether the mountain range has two peaks or two thousand peaks, its shape does not appear from nothing … mere differences in shape between the mountains are irrelevant to their dependence on some outside explanation (Rasmussen, 2019, p. 70-71).

Rasmussen’s point here is that differences in limits—shape, size, mass etc.—are  irrelevant towards being dependent—for  how does having more vertices or weight or size help with being independent and uncaused? If all limits are uniformly dependent, then this would make sense of why differences in limits are irrelevant.

I agree that differences in limits are irrelevant to being dependent, which is to say, it’s hard to see how a particular shape, size, belief, desire etc. could be the difference-maker with respect to being independent.  But I disagree that uniform dependence of limits is what accounts for the irrelevant differences.

What is the relevant difference that explains the difference between dependent things (that have an explanation) and the foundation (that has no outside cause or explanation)?  Nothing explains it, or so I say.

It seems to me we have everything we need in Rasmussen’s argument in chapter 3:

  1. A realm cannot be self-sufficient without any independent layer (because independence is the root of self-sufficiency).
    2. The blob of everything is a self-sufficient realm (because there is nothing beyond everything).
    3. Therefore, the blob of everything has an independent foundation (Rasmussen, 2019, p. 22).

Dependent things are explained by the things that cause them, and the independent (and necessary) foundation lacks an explanation because non-existent things can’t explain it, and there can’t be anything more foundational than the foundation to causally explain it, and there doesn’t seem to be anything to conceptually explain it either. I do not see the need to posit some further property that explains the difference between dependent and independent (and necessary) things, nor do I see how this further property would explain the difference, because everything seems to me to be an irrelevant difference. Take perfection again. As I indicated before, I think it’s the perfect-making properties (like independence) that explain perfection and not the other way around.


To repeat, I have only focused in on the arbitrary limits principle and have not addressed the rest of the book including the ‘construction problem’ and predictive capabilities of the perfect foundation hypothesis; that is a big topic in itself. Rasmussen gave three reasons for thinking the foundation lacks arbitrary limits: simplicity, explanatory depth, and uniformity. Concerning simplicity, I argued that the perfect foundation hypothesis is not as simple as it initially seems given that perfect-making properties are preconditions to perfection; that the intrinsic probability depends on the intrinsic probability of the cluster of perfect-making properties; and that the intrinsic probability of the perfect-making properties is inconclusive given an alternative method based on the principle of indifference.  Concerning explanatory depth, I gave a sketch of a reply that the perfect foundation hypothesis is too vague to be a sufficiently good explanation for limits. Concerning uniformity, I agreed that limits are irrelevant when it comes to explaining independence, but disagreed that we should think the foundation doesn’t have unexplained limits. I questioned whether we need to posit some property to explain the difference between dependent things and the foundation, and I questioned how any property could explain this difference.


Draper, Paul. “Simplicity and natural theology.” (2016).
Gwiazda, Jeremy. “Richard Swinburne, the existence of God, and principle P.” Sophia 48.4 (2009): 393.
Miller, Calum. “Is theism a simple hypothesis? The simplicity of omni-properties.” Religious Studies 52.1 (2016): 45-61.
Rasmussen, Joshua. How Reason Can Lead to God: A Philosopher’s Bridge to Faith. InterVarsity Press, 2019.
Rasmussen, Joshua, and Felipe Leon. Is God the Best Explanation of Things?: A Dialogue. Springer, 2019.
Plantinga, Alvin, and Michael Tooley. Knowledge of God. John Wiley & Sons, 2009.
Swinburne, Richard. The Existence of God. Oxford University Press on Demand, 2004.

[1] Rasmussen doesn’t exactly put it this way, but this is what I think follows from his view.

[2] For a defense of this view see Jeremy Koons’ Can God’s Goodness save the DCT from Euthyphro. Although he runs the argument based on goodness, a parallel argument can be applied to perfection. The idea is that if the perfect-making properties don’t explain perfection, then perfection becomes empty.

[3] Rasmussen has suggested to me that the order of explanation may depend on the sense of explanation (ontological, epistemic, conceptual, causal, etc.).

Posted in Josh Rasmussen, Philosophy of Religion | 1 Comment

Problems for Craig’s view on God and time

There are at least 3 ways to think of God’s eternality. First, God is everlasting in the sense that he is in time and exists from the infinite past and will exist into the infinite future. Second, God is timeless, outside of time altogether. Lastly, there’s Craig’s hybrid view: God is timeless sans creation and temporal since creation. In this post I criticize coherency of Craig’s hybrid view.

I can’t make sense of his view. It seems to me that if something it timeless, then it can’t become temporal, for if it became temporal the timeless part would come before the temporal part, which is a contradiction. Timeless things don’t come before anything; rather, timeless things stand apart from time altogether. As Leftow says:

If God is timeless, there is no before and after in His life. No phase of His life is earlier or later than any other phase, for only temporal durations and their phases stand in these relations. As it lacks earlier and later parts, an eternal life has no phases …. If God is timeless and a universe or time exists, then, there is no phase of His life during which He is without a universe or time, even if the universe or time had a beginning. For a life without phases cannot have one phase which is without the universe or time and another phase which is with it. If God is timeless, the whole of His life is identical with the ‘phase’ of it during which the universe or time exists, whether or not the universe or time began.

Craig disagrees:

But why could there not be two phases of God’s life, one atemporal and one temporal, which are not related to each other as earlier and later? Leftow merely assumes that if any phase of God’s life is timeless, the whole is timeless. But it may be the case that God’s atemporal phase does not exist temporally prior, technically speaking, to His temporal phase.


So if God is timeless, he is also unchanging, but it does not follow that He cannot change. I’d say that He can change and if He were to do so, He would cease to be timeless. And that’s exactly what I think He did. Whether God is timeless or temporal is a contingent property of God, dependent upon His will. What is impossible is changing while remaining timeless. But it seems to me that a timeless being can change and thereby cease to be timeless.

Consider an analogy Craig gives to support his view:

[A] man sitting changelessly from eternity could freely will to stand up; thus, a temporal effect arises from an eternally existing agent.

Craig holds to a relational view on time such that time only passes if there is change. By contrast, on a substantival view, time would pass in the absence of change. So I think Craig’s idea is that since the relational view is true, and the man is sitting there changelessly from eternity, he is timeless in his sitting state. The moment the man stands up, change occurs and he enters into time. (To make this situation analogous to God’s, we should think of the man as existing by himself, and his standing being the first change in the world.)

It seems to me the man’s sitting phase has to be temporally before his standing phase. Wasn’t he sitting temporally before he was standing? If that’s right, then he’s not timeless in his sitting phase but in time and at t=0. (This is analogous to God having an unchanging intention to create the universe, and then acting to create the universe.) If there isn’t a timeless phase of the man’s life then he (and God) begins to exist, according to Craig’s technical definition of begins to exist. Obviously Craig wouldn’t accept this. Craig’s move here is to distinguish between two senses of change, which I’ll call change_1 and change_2.

Craig explains change_1:

By an “event,” one means any change. Since any change takes time, there are no instantaneous events so defined. Neither could there be an infinitely slow event, since such an “event” would, in reality, be a changeless state. Therefore, any event will have a finite, nonzero duration.

So change_1 is an event that has a finite, nonzero duration, and this is the kind of change that cant regress infinitely, according to Craig. An event occurs when a thing gains or loses properties. This is the common sense understanding of change, and this is the sense I was thinking of when describing the man sitting from eternity. So the man would be sitting at t=0 and the change of him standing up took a finite, nonzero duration. Craig realizes that when God “changes” from being timeless to temporal it can’t be this sense of change, because this kind of change takes place over a finite, nonzero duration. So he introduces change_2:

… in the context of discussion of God’s relationship to time. Here I was using the word “change” in a different sense than the sense it carries in the kalam argument. In saying that God changed in creating the world, I meant merely that God is not the same in His timeless state and in His first temporal state. He has different properties (my italics). In that sense He changes. But that difference does not take place over time and therefore is not a change or an event in the sense spoken of in the kalam argument.

So when God changes from being timeless to temporal, it’s of this change_2 kind: God changes from being timeless to temporal because being timeless is a different property than being temporal. It seems Craig’s words are misleading when he says that God enters into time and ceases to be timeless. This kind of entering and ceasing is merely the having of different properties.

If God changes from being timeless to temporal in this change_2 sense, this would avoid the contradiction I stated earlier, but it introduces a new problem. If God is not changing in the change_1 sense, it’s hard to see how God is not still timeless. If God is still timeless, and given that Craig thinks that God is temporal since creation, then God is both timeless and temporal right now, a contradiction. It’s change_1 that’s needed for God to lose the property of timelessness (which is itself contradictory for it would mean he’s timeless in the past). So it seems to me that if God is timeless, God is timeless through and through.

One might think that there is no contradiction in God being both timeless and temporal if we qualify it somehow. Let’s consider two ways. First, let’s consider the sitting man again. Perhaps as the man stands his foot remains unchanging while the rest of his body moves such that his foot is timeless while the rest of his body enters time; so there is a part of the man that is timeless and another part that is temporal. Contradiction avoided. This won’t work on Craig’s view because any change extrinsic to the foot is enough to bring the foot into time. Craig explains:

Let’s imagine this hypothetical rock that is absolutely changeless and isolated in outer space. Then imagine that a meteor whizzes by and another meteor whizzes by. Clearly the rock would not be timeless even though it is intrinsically changeless. Why? Because it changes in its relation to other changing things about it. First there was the one meteor going by, then later another meteor went by. The rock, though changeless intrinsically, would clearly be in time because it is related to changing things. Since God is really related to a changing temporal world, God would undergo extrinsic change and therefore he would be in time. This seems to me to be a very powerful argument for God’s being temporal.

So even if the foot is changeless, the changing of the rest of the body is enough to bring the foot into time. Even if Craig is wrong here and the foot is timeless, I don’t see how this could help the holder of Craig’s hybrid view, for God doesn’t have any internal parts that could play the role of the foot.

Let’s look at a second way to qualify the view to avoid the contradiction. Consider the two propostions:
A. Trump won the U.S. presidential election.
B. Trump did not win the U.S. presidential election.

Unqualified, these two propositions contradict each other. Now suppose we add a temporal index.

A’. Trump won the U.S. presidential election in 2016.
B’. Trump did not win the U.S. presidential election in 2012.

Now the contradiction has been avoided. Craig does something similar when he states that God is timeless sans creation and temporal since creation. Craig says, “God’s state of existing timelessly sans creation can serve logically as a sort of temporal index.” With this qualification, it’s not immediately obvious that Craig’s hybrid view is contradictory. Craig supports his hybrid view with a though experiment:

The impression that the state of affairs of God existing changelessly sans creation is timeless may be reinforced by a thought experiment: think of God in a changeless, solitary state in a possible world W* in which He freely refrains from creation. In such a world, it is entirely plausible and coherent to conceive of such a state as timeless. But no intrinsic difference exists between such a state and the state of affairs of God existing sans creation in the actual world. The allegedly initial segment of the actual world TW is perfectly similar to the world W*. It seems groundless to say that in one world God is temporal in such a state and in the other world atemporal.

Suppose there is no intrinsic difference in God (sans creation) between the two worlds, as Craig says. I don’t see how this helps, for according to Craig intrinsic or extrinsic change is enough for time to pass, and we have both intrinsic and extrinsic change in God in the actual world. Recall Craig’s unchanging rock with a meteor whizzing by. Craig thinks that this extrinsic change is enough for the rock to be in time. We could run a parallel thought experiment. Suppose in w1 a solitary unchanging rock exists, and in w2 that same unchanging rock exists but a meteor whizzes by. Since there is no intrinsic difference in the rock between the two worlds, should we say the rock is timeless in w2? No. Craig’s thought experiment says it is in time. So I think his rock thought experiment and his intrinsic-similarity thought experiment are in tension.

Ultimately, I’m skeptical if Craig’s “sort of temporal index” is legitimate. I don’t think there is state of affairs of God existing sans creation in the actual world. Even if there is some possible world with the state of affairs of God existing alone sans creation, this world is not like that world. There never is an obtaining state of affairs of the universe not existing, and this is true even assuming God caused the universe, an A-theory of time, and that time had a beginning. If you disagree, then when did the universe not exist? It seems to me that God’s being timeless sans the universe in the actual world is merely conceptual. For example, imagine an ball resting on a cushion from past infinity. We can conceptualize the ball sans cushion, but in reality there is no obtaining state of affairs of the ball sans cushion. We could similarly say that the unchanging rock is timeless sans whizzing meteor in the world where the meteor whizzes by, and this would also be merely conceptual when in reality both the rock and meteor co-exist.

If I’m right and Craig’s hybrid view is incoherent, then the theist needs to adopt a different model of God’s relationship to time. (I leave Padgett and Swinburne’s view aside. Craig thinks that view reduces to God being past infinite.) On the one hand, the theist could think that God is everlasting and give up the idea that an infinite past is metaphysically impossible. He could still retain the Kalam, but he’d have to give up the philosophical argument against an infinite past. On the other hand, the theist could think that God is timeless simpliciter, like many classical theists do. This theist will have to face Craig’s argument that extrinsic change is enough to bring God into time. (There are other issues with a timeless, omniscient God ability to know tensed facts, which I’ll leave aside.) Lastly, and unlikely, if these two alternatives are too much to stomach, the theist could hold that God begins to exist, and question the idea that everything that begins to exist has a cause.


Leftow, Brian. “Why didn’t God Create the World Sooner?.” Religious studies 27.2 (1991): 157-172.

Posted in Philosophy of Religion | Tagged , , | 3 Comments

The epistemology of modality and some metaphysical consequences

I want to consider how we know what worlds are metaphysically possible. First, let’s state what a possible world is. I like to think of possible worlds in the way Kripke states it: “‘Possible worlds’ are total ‘ways the world might have been’, or states or histories of the entire world (18).” So here’s the question: How do we know the ways the world might have been, or might be? Here I’m ignoring any abstract reality and restricting the question to causal reality.

One popular view on the epistemology of modality uses a principle that says conceivability is a guide to possibility—let’s call it CGP. We can see this principle in Hume’s Treatise:

‘Tis an established maxim in metaphysics, That whatever the mind clearly conceives includes the idea of possible existence, or in other words, that nothing we imagine is absolutely impossible. We can form the idea of a golden mountain, and from thence conclude that such a mountain may actually exist. We can form no idea of a mountain without a valley, and therefore regard it as impossible. (T 1.2.2)

The idea is that if we can conceive of something in some robust way, then that tells us that the world could have turned out that way. One counterexample to CGP is Goldbach’s conjecture: we can conceive of it being true or false, but it is necessarily one or the other. I think Chalmers’ more robust version of conceivability can bypass this objection. Another counterexample is Kripke’s a posteriori necessities involving natural kinds: e.g. we can conceive of water not being H₂O but their identity is necessary. For the purposes of this post, I’ll ignore Kripke since I think there are bigger problems with CGP. I’m concerned with the idea that by conceiving of a different coherent set of laws, we can know that the actual laws of nature could have been different.

The main problem with CGP is that there is no reason to think that a priori conceivability has anything to do with ways the world might have been. Granted things have to be logically possible for things to be metaphysically possible, but defenders of CGP are making a bolder claim. Suppose that we can coherently conceive of some laws of physics where things travel faster than the speed of light. How does that show us that the actual world could have turned out that way? Where’s the link? I suspect that defenders of the view take it as a brute intuition. A priori reasoning, while being indispensable in some areas, can only take us so far. Hegel went too far when he argued a priori that there cannot be more than six planets. Similarly, the CGP goes too far in my estimation.

Since I’ve ruled out a priori conceivability, and I’m not a skeptic about modal epistemology, the other option is that we can know what’s possible a posteriori. So how to we figure out our modal landscape? First, we know trivially that the actual world is possible, and we know what the actual world contains from experience. Second, if the laws of nature are indeterministic, that brings about additional possible ways that the world might have been. Of course, we find out about the laws of nature empirically. So the trick to figuring out “ways the world might have been” is to look at the actual world and figure out the laws of nature. If determinism is true, then there is only one possible world; if indeterminism is true, we have many possible worlds.

There’s an interesting result once you ground possibilities in the laws of nature in the actual world. It turns out that the fundamental laws of nature are necessary; they couldn’t have been different. Why? Because if possibilities come from the fundamental laws of nature (more specifically, the causal powers of actually existing things), and for the fundamental laws of nature to possibly be different, there would have to be even more fundamental laws to allow for that possibility, which, by definition, is impossible.

Graham Oppy makes an interesting observation about a view like this:

My favourite theory of modality has the evident advantage of theoretical frugality. On the one hand, if there are objective chances, then any theory of modality is surely committed to the possibility of the outcomes that lie in the relevant objective chance distributions. On the other hand, it is not clear that we have good reason to commit ourselves to any possibilities beyond those that are required by whatever objective chances there might be; at the very least, any expansion of the range of possibilities clearly requires some kind of justification. (47)

This theory has the advantage of theoretical frugality. Those who think that chances (or possibilities) can come somewhere besides the laws of nature owe us some kind of justification. If they think there are chances that do not come from any actually existing things, they are positing a sui generis chance—a chance from nowhere. Parsimony dictates that we not accept mysterious things without reason.

This theory also has the advantage of not falling to the “relevance problem.” If possibility is grounded in Plantingean propositions or Lewisian concrete worlds, what relevance do they have to how the actual world could have been? As William Lycan puts it:

…why should we suppose that real possibility and other modalities in this world have anything to do with specially configured sets of items, whether sentences or propositions or matter-elements? It seems unlikely that what fundamentally makes it true that there could have been talking donkeys is that there exists a fabulously complex set of some sort. (Lycan 1998, 92)

By contrast, if we ground possibility in the actual laws of nature, it has the advantage of being eminently relevant to how the actual world could have been.

Chalmers, David. “Does conceivability entail possibility?.” Conceivability and possibility (2002): 145-200.
Goldschmidt, Tyron, ed. The Puzzle of Existence: Why is There Something Rather Than Nothing?. Routledge, 2014.
Hume, David. A treatise of human nature. Courier Corporation, 2003.
Kripke, Saul A. “Naming and necessity.” Semantics of natural language. Springer, Dordrecht, 1972.
Lycan, William. 1998. “Possible Worlds and Possibilia:’ In Contemporary Readings in the Foundations of Metaphysics, Stephen Laurence and Cynthia Macdonald (eds). Oxford: Blackwell.

Posted in Epistemology, Metaphysics | Tagged , | Leave a comment