Carl Sagan: Where Did God Come From?

Watch this:



"Where did God come from?" This is a popular objection from atheists. Unfortunately, it's also completely misguided, and only exposes the atheist for being utterly unfamiliar with classic monotheism.

To explain…

Starting with Aristotle, which is where classic monotheism begins to really gel, the basic idea is that all changeable things (such as birds, people, planets, stars) cannot collectively be responsible for their own existence, because then they would be causing themselves to exist, and so would have to exist prior to them existing, which is a logical contradiction. To solve this, Aristotle hypothesizes that there must be something unchangeable which is responsible for the existence of changeable things. Because it is unchangeable, it cannot have a cause even in principle. A cause implies a change, so if something is unchangeable then it doesn't need nor can it even have a cause.

It must be emphasized that Aristotle is not speaking of the thing that triggered the Big Bang, but rather of a "substrate" that grounds the existence of changeable things, even if the universe is eternally old. Just like a room full of dancing people, even if infinitely old, still has to have a floor in order to hold the people up.

So now hopefully it is clear why Sagan's objection makes no sense, and is in fact complete gibberish. Asking where God came from is asking about a change in God, that he came from somewhere, or was caused by something. But the whole point of classic monotheism is that God is the unchangeable reality that grounds the existence of changeable things. So Sagan's objection translates as follows:

• What changed the thing that is not changeable?

You can see that this is a nonsense question. It entails a contradiction. It's like asking:

• Which mountain is higher than the highest mountain?

Peter Boghossian's "A Manual For Creating Atheists"

Instead of a full review of Peter Boghossian's new book "A Manual For Creating Atheists", I figured I would just focus on one tiny section that relates to my specialty: arguments for the existence of God.

Boghossian aims to create a full method of talking people out of "faith", by which he seems to mean "theism". His technique involves learning the Socratic method and being nice and gentle to believers, unlike, say, Dawkins. I leave my opinion of this idea to the side for now, but it strikes me as being very….I don't know. Evangelical. Like…"Jehovah's Atheists" or something. Proselytizers. The flip side of the apologist coin. It makes me think of the various times in history where rebels fought against a communist dictator only to end up becoming…communist dictators.

But I digress. My focus will only be on his treatment of cosmological, ontological, and teleological arguments. Since the evidence for the existence of God will generally be in the form of these arguments, I want to see what kind of objections he offers.

Cosmological Arguments

He starts off in Chapter X: Apologetics 101. First up is what appears to be the cosmological argument. First, he quotes Ray Comfort, a Christian apologist:

“Why is there something rather than nothing? You have faith that there was no Creator.” “Bear in mind that an atheist believes that all these miraculous coincidences took place by chance. But he doesn’t just believe that man and woman came into being without a Creator, but that all of creation did—amazing flowers, massive trees, succulent fruits, beautiful birds, the animal kingdom, the sea, fish, natural laws, etc. His faith is much greater than mine.” —Ray Comfort, You Can Lead an Atheist to Evidence, but You Can’t Make Him Think (2009, p. 2) 

And then he says:

This is the best argument I’ve heard for the existence of God. 

OK, let me stop right here. An argument from Ray Comfort is the best argument he's ever heard??!! Ray Comfort?! Ray Comfort is one of the worst defenders of Christianity the world has ever seen. He is infamous for using a banana to prove God's existence. He is an evolution-denier as well. This is like playing chess against a newborn and then claiming that this was the best chess opponent you've ever seen. Has Boghossian never heard of Plato, Aristotle, Maimonides, Aquinas, Leibniz, or Richard Swinburne?!

But I'll remain charitable. Let's just say that he addressing the cosmological argument in general:

It’s the trump card played by believers. However, it doesn’t work. There are several related ways to respond to why there’s something rather than nothing: “Why assume nothing is the default?” This is a question that has no answer. As prolific German philosopher Adolf Grünbaum states, “Why be astonished at being at all? To marvel at existence is to assume that nothingness is somehow more natural, more restful. But why? The ancients started with matter, not the void; perhaps nothingness is stranger than being” (Holt, 2012). 

I could retort here that existence needs explaining. That I don't need to explain why there isn't an elephant in my bedroom right now, but if there were one, I would need to explain why there is.

But even that isn't necessary. One of the major cosmological arguments simply says that contingent things need an explanation. "Contingent" means that it could exist, or not, logically speaking. Most things we are familiar with are contingent. Your parents might never have met. The Earth might not have formed. Mt. Everest might not have formed. These things exist, but might not have. They are contingent. And so they stand in need of an explanation for their existence. The argument could be put like this:

1. All contingents have an explanation

2. The set of all contingents is itself contingent

3. Therefore, the set of all contingents has an explanation

And that explanation cannot be something contingent, because then it would be part of the set we are trying to explain. So it must be something outside the set, i.e. something non-contingent. Something non-contingent would be logically necessary, and hence capable of existing in any possible alternate world, including ones in which no matter exists, or no space, or energy. So something non-contingent must be spaceless, timeless, immaterial, and capable of explaining the existence of contingent things. I would challenge Peter Boghossian to show which premise is false in this argument. Premise 1 is well-supported by science, and we never see a counter example. And premise 2 follows logically from what it means to be contingent: if one member of the set could not exist, and two members of the set could not exist, and all members of the set could not exist, then the set itself could not exist as well, and is therefore contingent. 

Continuing:

Similarly, “How do you know the universe didn’t always exist?” Even if appeals are made to the Big Bang, one can never know either that reality is one endless time loop with Big Bangs strung together for eternity, or that à la American theoretical physicist Brian Greene, we’re part of a larger multiverse with an infinite number of Big Bangs constantly occurring. 

Note that the argument above says nothing about the beginning of the universe. Perhaps the universe is eternally old. Nonetheless, all contingents have an explanation for their existence…etc. Answering "it always existed" tells me how old something is, but does not tell me the explanation for its existence. "Why is the hammer in the fridge?" "It's always been in there." "OK, but it could have always been on the counter. What is the explanation for it having always existed in the fridge rather than always on the counter?"

Richard Taylor, in his book Metaphysics, provides an interesting breakdown of this contingency argument. He argues that if you were to come across a translucent sphere in the forest, you would want to explain its existence. And saying that it always existed in the forest does not offer an explanation. Why an eternally existing translucent ball rather than an eternally existing blue one? 

And even if we do accept by fiat, given our limited knowledge, that something rather than nothing is extraordinary, does that give license to make up answers as to why this is the case? It begs the question: is it better to pretend we know an answer to something we don’t actually know, or is it better to simply be honest and say, “I don’t know?”

Of course we should not just make up answers. But if the contingency argument is sound, and a necessary something-or-other exists, then you can extract at least some properties from it that line up with what has traditionally been thought of as "God", such as eternality, spacelessness, timelessness, immateriality, power, etc. Those attributes automatically drop out of the necessary, non-contingent something-or-other that is the conclusion of the contingency argument. They aren't "made up".

But we can go even farther than that. Take Aquinas's cosmological argument. Like the contingency argument, it does not require that the universe began to exist. Rather, what it argues is that any particular object is dependent on further states of affairs for its ongoing existence from moment to moment. For example, a liquid lake is dependent upon warm air for its moment-to-moment existence. But the warm air is also dependent upon further, current, factors, such as atmospheric pressure and heat from the Sun. But those states of affairs are dependent upon yet further states, and so on. This chain has to terminate, not in the Big Bang, but in a present state of affairs that is not dependent upon any further states of affairs. If it were dependent on something further, then it just wouldn't be the termination point in the first place. And once Aquinas gets to something not dependent upon anything further, something that is just subsistent existence itself. Something maxed out. And from that, he quite easily argues for the typical divine attributes: since it is maxed out, it must be all-powerful, all-knowing, etc. In fact, he goes into great detail on each of these attributes and answers objections.

The possibility that the universe always existed cannot be ruled out. This by definition casts doubt on a creator. No faith is needed to posit that the universe may have always existed

Again, see above for two cosmological arguments that do not require a beginning to the universe. In fact, of about a dozen or so cosmological arguments defended by the major figures of philosophy, only two of them argues that the universe must have had a beginning. Specifically, the arguments from al Kindi and al Ghazali argue for a beginning to the universe. None of the others, from Plato, Aristotle, Plotinus, Aquinas, or Leibniz require the universe to have a beginning.

And there's more. Aristotle's famous unmoved mover, for example. Aristotle first begins by proving that the universe must be eternally old by proving that change could never have started, because if it did, then something would have had to start it and thus would have been a change itself, and so something else must have started that, and so on ad infinitum. So his first step is to argue that change eternally existed, and will always exist. And from that, he argues that not everything can be a changeable changer, because then change could stop. There must be something that is an unchangeable changer (an unmoved mover) that continually causes change without ever stopping, to keep in line with the first premise that change is eternal.

And yet more. Consider Plotinus's argument for The One. He argues that most objects are composed of parts, which are themselves composed of further parts, and so forth. A human being is composed of organs, which are composed of cells, which are composed of molecules, and so on. The simplest constituent of the universe must therefore not be composed of parts, and must be unchangeable. It is the simplest thing in the universe, ineffable, and beyond being or non-being. Plotinus calls it The One.

So you can clearly see that even if the universe is eternally old, this does not touch the cosmological argument in 10 out of its 12 versions. So merely stating that the universe could be infinitely old does not come close to refuting the idea of a creator.

Ontological Arguments

Now, Boghossian starts out with what at face value is an awful argument for God's existence and could easily be seen to be appeal to ignorance (you can't prove not-X, therefore X):

The basic idea is that because you can’t prove that there’s not a God, then God must exist. Of all of the defenses of faith, it is most difficult to comprehend how someone could actually offer this as a legitimate defense for faith or for belief in God.

As worded, he is correct that this is a terrible argument. Boghossian challenges his interlocutors with "little blue men on Venus", saying that there is no evidence for their existence and that they must therefore exist. He continues:

Most people will get the point and then say there’s something different about God. That is, this line of argument works against everything except God. (Here I’m reminded of defenders of Anselm’s argument for the existence of God. Every time someone would bring up an objection, they’d state that the argument only works with God.) When the respondent says there is something special about God that makes this argument not work, then I always press them to know what’s different about God. I’ve yet to hear a coherent answer to this question.

The difference is that God is, as postulated, is a non-contingent, or necessary, being. This argument would work with anything that is necessary, such as mathematical propositions. And from modal logic, there cannot be a merely possible necessary being.

Let's look a little closer. The word "possible" means something could be true, or not. It isn't logically impossible. It is possible that JFK was not assassinated, even though he was. It is possible that he was assassinated, and he was. Both situations are possible. The word "necessary" means something must be true and cannot logically be false. It is necessary that  2 + 2 = 4. It could not have been otherwise. The word "impossible" means that something is logically impossible. It could not have happened. It is impossible that a bachelor is married. Since "bachelor" means "unmarried man", it is logically contradictory for someone to be both married and not married. I.e., be a married bachelor.

These are called "modal claims". One useful tool for examining modal claims is called "possible worlds". These are not actually existent worlds, but just concepts to examine what we mean by "possible" and "necessary." Think of them as alternate realities. The way the world could have been.

So we would say that there are possible worlds where JFK was not assassinated. And there are no possible worlds where there is a married bachelor. And 2 + 2 = 4 in all possible worlds. So the word "necessary" translates to "true in all possible worlds." And the word "possible" translates to "true in some possible worlds, and false in others."

Now, when dealing with a necessary being, it is contradictory to say that that necessary being is merely possible, because you would in effect be saying that it exists in A) all possible worlds, and B) some possible worlds. This is contradictory. Something cannot simultaneously exist in all possible worlds, and not all possible worlds. In other words, there cannot be a merely possible necessary being.

In short, if something is possibly necessary, then it is necessary. If it is possible that a necessary being exists, then by definition it exists in at least one possible world. But, being necessary, then it exists in all possible worlds, which of course includes the actual one. The argument could be written like this:

1. If it is possible there is a necessary being, then there is a necessary being
2. It is possible there is a necessary being
3. Therefore, there is a necessary being

Premise 2 is the point at which the argument hinges, and is the focus of the argument Boghossian gives above. A theist could say that since no argument has been given to show that God, a non-contingent (necessary) being, is logically impossible (i.e. does not exist), then it follows that God is at least possible, in which case premise 2 is confirmed and the argument goes through.

Now, one might object that just because it has not been shown that God is impossible, it does not follow that God is possible. And that is correct. However, a theist could use it as at least some inductive support for the premise. If one cannot find a prima facie contradiction in the concept of God, and atheists have spent 2500 years trying to do so without unqualified success, then a theist could use that to argue that there is no contradiction, otherwise it would have been found by now. Or at least, the burden is definitely on the atheist at that point.

In short, the theist could say: "You have not proven that God does not exist due to any logical contradiction, and therefore God exists in some possible world, and therefore, because he is a necessary being, in all possible worlds. Or to put it another way, you have not proven God is impossible, therefore God exists."

So the apparently terrible argument from ignorance that was made by the theist could be interpreted as a perfectly logical modal ontological argument. Now, even if the theist didn't mean it this way, it still behooves us to try to interpret our interlocutors in the best possible way, even going so far as to attributing to them a better argument they could have made but did not. This is basic principle of charity 101.

Conclusion

This is all I could discern of the main arguments for the existence of God in Boghossian's book. Were I a theist, I would be one because of evidence such as Aristotle's cosmological argument, or Plotinus's. I would not be a theist due to "faith", interpreted as "belief in something without evidence." I would be a theist because I would agree that contingent objects are dependent on further factors, and that there must therefore be some non-contingent reality explaining their existence. I would believe because there is evidence. If I were a theist, Boghossian's book would not put a dent in my belief. The book seems to be aimed at evangelicals and fundamentalists as found in the Southern U.S., not at any academic or thoughtful theist.

Richard Dawkins vs Thomas Aquinas

Of the "new atheists", Richard Dawkins is one of the few who seems to realize that he ought to address the arguments theists have given for theism. Harris, Hitchens, and Dennet don't really address theistic arguments at all in their most popular books. So credit due where credit is due. However, when Dawkins addresses the Thomistic First Cause arguments, he mucks them up so badly that they are beyond recognizable. Let's take a closer look at what Dawkins says about Aquinas in The God Delusion:

All three of these arguments rely upon the idea of a regress and invoke God to terminate it. They make the entirely unwarranted assumption that God himself is immune to the regress.

The arguments in question do not make an unwarranted assumption at all. The whole point is to argue that a "vertical" series of in esse causes must terminate in something that can "hold up the whole stack" without itself needing to be held up by anything further (for more detail, see here). We use the same kind of reasoning when we are listening to the radio and we wonder where the radio station is transmitting from: our radio is a receiver, which entails that there must be a transmitter, or something that can give the radio signal without having to get it from anything further. Or how a series of spring-loaded clamps, each dependent on the next clamp for being held shut, must terminate in a clamp that doesn't need to be held shut by any further clamps. As you can see, this is not unwarranted at all.

Next, Dawkins talks about the divine attributes:

Even if we allow the dubious luxury of arbitrarily conjuring up a terminator to an infinite regress and giving it a name, simply because we need one, there is absolutely no reason to endow that terminator with any of the properties normally ascribed to God: omnipotence, omniscience, goodness, creativity of design...

Did he miss questions 3 through 26 of Aquinas's Summa Theologica? Once Aquinas argues for a first cause, he then goes on to extract the divine attributes of such a thing one by one, such as simplicity, omniscience, omnipotence, and even lovingness and mercy.. For a shorter look, see my article on the attributes of pure actuality. Dawkins seems completely unaware of this.

Incidentally, it has not escaped the notice of logicians that omniscience and omnipotence are mutually incompatible. If God is omniscient, he must already know how he is going to intervene to change the course of history using his omnipotence. But that means he can't change his mind about his intervention, which means he is not omnipotent.

The whole point of the argument is to say that all changeable things must trace "down" to an unchangeable bottom-level. As something unchangeable, the bottom-level does not live "in time" like the rest of us, but its existence is simultaneously whole. There is no "future" for it. Its actions are already in place. So it makes no sense to talk about it changing its mind, since everything is already done, from its perspective. Had Dawkins been knowledgeable about the argument, he would not have made this mistake. Continuing:

To return to the infinite regress and the futility of invoking God to terminate it, it is more parsimonious to conjure up, say, a 'big bang singularity', or some other physical concept as yet unknown. Calling it God is at best unhelpful and at worst perniciously misleading.

And here Dawkins makes clear that he thinks, as most people do, that Aquinas is arguing for a beginning to the universe. The argument that the universe had a beginning is called the Kalam cosmological argument, and Aquinas clearly rejects this argument because he does not think it can be proven philosophically that the universe had a beginning. To quote Aquinas: "By faith alone do we hold, and by no demonstration can it be proven, that the universe did not always exist..."

Aquinas's argument is for the most fundamental layer of reality in the present, not the event that kickstarted the Big Bang. If you don't understand the difference between in fieri and in esse causes, then you cannot understand the argument. Dawkins, as most people, thinks Aquinas is talking about "in fieri" causes, stretching back to the Big Bang, rather than "in esse" causes stretching down in the present. 

Some regresses do reach a natural terminator. Scientists used to wonder what would happen if you could dissect, say, gold into the smallest possible pieces. Why shouldn't you cut one of those pieces in half and produce an even smaller smidgen of gold? The regress in this case is decisively terminated by the atom. The smallest possible piece of gold is a nucleus consisting of exactly seventy-nine protons and a slightly larger number of neutrons, attended by a swarm of seventy-nine electrons. If you 'cut' gold any further than the level of the single atom, whatever else you get it is not gold. The atom provides a natural terminator to the Crumboblious Cutlets type of regress. It is by no means clear that God provides a natural terminator to the regresses of Aquinas.

Good! I'm glad he is open to the idea of some regresses having a natural terminator. Because a vertical causal series does have a terminator. Dawkins would know this if he were aware that the argument is for an in esse chain of sustaining causes, rather than an in fieri chain of events. But, of course, he doesn't know this and so can't think of why a chain of events would necessarily need a first cause, as even Aquinas himself agrees! As Aquinas says: " ....it is not impossible to proceed to infinity "accidentally" as regards efficient causes...as an artificer acts by means of many hammers accidentally, because one after the other may be broken. It is accidental, therefore, that one particular hammer acts after the action of another..." That is to say, a chain of events could stretch back into the past for infinity, and does not necessarily need a first cause. It is only a concurrent chain of sustaining causes that has an natural terminator.

Hopefully, this will show that Dawkins doesn't get Aquinas even close to right, and if anyone uses Dawkins to "refute" him, they are only exposing their own confirmation biases.

Feser's "The Road From Atheism"

I thought I would summarize the main points of Edward Feser's conversion from atheism to theism. I may explore these in more detail point-by-point over time.

• Atheism caused not by the argument from evil, but by the evidentialist objection: there is no (good) evidence for the existence of God.

• The difference between the theistic personalism of modern evangelicals etc, and classical theism. He still thinks theistic personalism (God as a magical sky wizard) is implausible. Classical theism (God as existence itself) is more plausible.

• The content of propositions are not identifiable with the sentences that express it, or dispositions, or psychological states, etc (i.e., some form of Platonic realism is true).

• Materialism of the mind is hopeless (John Searle's Rediscovery of the Mind)

• Russell's view that matter is more problematic than mind, because we are intimately familiar with mind but not matter, and...

• ...physics cannot be the whole story, because physics gives you only the abstract structure of matter

• Mind cannot be assimilated to matter if one wants to avoid panpsychism.

• Naturalist theories of "aboutness" are impossible

• "Classical" problems in philosophy are caused by anti-Aristotelianism, and most modern atheist moves presuppose these assumptions

• Aristotelianism seems much more plausible, especially as a solution to all the "classical" philosophical problems.

• Leibnizian cosmological arguments that say the universe must have an explanation. To disagree with this is less plausible and more rhetorical than it is well-argued.

• Most philosophers are way off base when treating Aquinas' arguments for the existence of God.

Is Atheism Just A "Lack of Belief"?

Often, atheists will claim that atheism is not a positive claim, but rather a lack of belief in gods. Such a position is merely negative, and doesn't tell us what the person actually thinks about the God question. This can be shown by looking at the possible positions one can take when one hears about gods:

1. Gods exist
2. Gods do not exist
3. I don't know
4. No one knows
5. I don't care
6. Etc.

By saying that they "merely" lack belief, they are saying that they are not 1.

OK, that tells us what the atheist is not, but not what he is. By saying "not 1", the atheist still needs to clarify which one he is, 2 through 6. But if he would just tell us which one he is right off the bat, that would then entail not 1, and so there is no reason to start with not 1.

Evidential Problem of Evil

Evidential Problem of Evil

This argument attempts to decrease the probability of the existence of the God of classical theism, unlike the logical problem of evil which attempts to refute it.

William Rowe's formulation

One of the more influential versions of the argument, first published in 1978.

(1) God would prevent suffering, unless he could not do so without losing a greater good

(2) There exist examples of suffering that God could have prevented without losing some greater good

(3) Therefore, God does not exist

Premise 1

Support for premise 1 is appeal to traditional notions of God as all-good. Most theists agree with premise 1.

However, open theism is opposed to premise 1. Open theism is the view that God's omniscience does not include foreknowledge (knowing what free agents other than God will do) or middle knowledge (knowing what free agents would have alternately done in any situation). But open theism advocates a God that is not as involved in human affairs, since he does not know the future and is playing dice with history. In this sense, open theism is more like deism and less compatible with the classical definition of God.

Premise 2

Most critics focus on premise 2 of the argument. Rowe supports premise 2 with appeals to two examples of apparently "gratuitous" evil (evil that serves no greater good):

Evil 1: A deer trapped under a log in a forest fire, burned and slowly dying for days

Evil 2: A real-life example of a little girl who was raped and beaten to death by her mother's ex-boyfriend.

From these examples, Rowe asserts the following inference:

P: We do not know of any greater good that could arise from these evils

Q: Therefore, there is no greater good that could arise from these evils

Rowe includes among "greater goods" anything that we can conceive of, past or future. Even "eternal bliss" is an acceptable greater good according to Rowe. The inference from P to Q is inductive, but Rowe defends it as rational. I.e., the truth of P is taken to be strong evidential grounds for the truth of Q.

The Skeptical Theist response

The traditional concept of God is that his ways are unknowable to man, and hence this presents problems for the inference from P to Q. Wykstra developed this response in 1984, known as CORNEA.

Wykstra's CORNEA Critique

Stephen Wykstra's Condition of Reasonable Epistemic Access (CORNEA) says the following:

C: A human is entitled to infer that [there is no x] from [so far as I can tell, there is no x] only if it is reasonable for that human to believe that if there were an x, it is likely that she would perceive it.

Wykstra calls this the "no-seeum assumption". This principle appears unobjectionable at first sight. Example: If someone looks at a garden from a distance and is unable to perceive any caterpillars in it, it is not reasonable for them to then assume that there aren't any.

Wykstra says that Rowe's inference from P to Q is a "no-seeum assumption." That if there is some greater good that will come about by allowing evils of the above type, it is unlikely human beings would be able to discern them.

Wykstra's parent analogy

Wykstra makes the analogy of a parent who allows her children to experience pains for greater goods (e.g. vaccination), and that a parent's cognitive abilities compared with her baby are at least comparable to human cognitivie abilities compared to God. So it is just as unlikely that we would be able to percieve God's reasons for allowing evils as it would be for a baby to perceive the reasons for a vaccination: not likely at all.

Alston's analogies

Wyksta's analogy is open to criticism, so Alston came up with two different ones:

Student: A student may not understand the reasons an expert does something, but from this he cannot infer that there are no reasons.

Scope of knowledge: If the field of knowledge in a subject is unknown in scope. I.e., because we can't find aliens, we cannot infer that there are no aliens out there.

Based on the above analogies, Alston concludes that Rowe's inference from P to Q is a "no-seeum inference" and hence invalid. However, Rowe responds that unlike in the two analogies, an all-good being should either tell us his reasons for permitting suffering, or at least make his presence more obvious.

Building a theodicy

A theodicy is an attempt to provide a plausible reason for God permitting suffering. This is distinguished from a "defense," which is used against the logical problem of evil by trying to show a possible reason that God allows suffering, even if that reason might be implausible.

Sketch of a theodicy

A theodicy might make use of several tactics. For instance:

Soul making: the idea that God uses pain and suffering for the betterment of character; a person partaking in only hedonistic pleasures non-stop will tend to have a bad character.

Free will: the idea that God allows morally significant free will so that human beings are moral agents and not just robots programmed to act a certain way.

Heavenly bliss: the idea that after death a soul will go to heaven and experience eternal happiness.

So, for instance, the above evils might be explained by serious misuse of free will, soul-building for the people left alive, and eternal heavenly bliss for those who die.

Objections: Dispute that free will exists; we have a duty to curtail someone's free will if they misuse it, and so Go should do the same; natural evil is difficult to account for with theodicies.

Responses to natural evil: Natural evil could be part of soul-building (Hick); it could be to give us knowledge of evil so that we know we know what moral choices are (Swinburne).

Further responses to the evidential problem of evil

If theists think the arguments for the existence of God are strong enough to offset the evidential problem of evil, they can turn the argument around to say that since an all good God exists, then gratuitous evil (evil that serves no greater good) does not exist. Rowe calls this the "G E Moore shift." The formal argument:

(1) God would prevent suffering, unless he could not do so without losing a greater good

(2) God exists

(3) Therefore, there is no suffering that God could have prevented without losing some greater good

Rowe says that this is the theists' best response. However, the problem is that the typical arguments for God (cosmological, teleological) do not seek to establish an all good being, and thus the theist faces severe difficulty in offsetting the obvious suffering in the world with arguments for an all good being.

Logical Problem of Evil

The Problem

The God of classical theism is defined as omnipotent, omniscient, and all good. But this conflicts with the existence of pain and suffering in the world, which indicates that God is either not omnipotent, not omniscient, or not all good. The formal argument:

(1) God is omniscient, omnipotent, and all good (from the classical definition of theism)

(2) Evil exists in the world

(3) If God is unable to prevent evil, then he is not omnipotent

(4) If God is unaware of evil, then he is not omniscient

(5) If God is unwilling to prevent evil, then he is not all good

(6) Therefore, God is either not omniscient, not omnipotent, or not good

(7) Both 1 and 6 are contradictory

(8) Therefore, the God of classical theism does not exist

Attempts to resolve this problem results in a lesser God than theists would be willing to accept. The problem of evil as a logical problem was primarily defended by J.L. Mackie in the mid-20th Century.

Logical Consistency

Logical consistency requires that it be possible for several statements to all be true at the same time. If it can be shown that there is a possible way of reconciling them, even if implausible or unlikely, then the statements are not logically contradictory.

Logical consistency and the logical problem of evil

If it is possible that God has a morally sufficient reason to permit evil, then the logical contradiction does not hold. The conclusion would then be that either God does not exist or that he has a morally sufficient reason to permit evil. An example would be of a child receiving a vaccination. So long as it's even possible that God has long-term, morally sufficient reasons for permitting evil, there is no logical contradiction between an all good God and the existence of evil.

Plantinga's Free Will Defense

Alvin Plantinga makes a suggestion for one possible morally sufficient reason that God would have for permitting evil: morally significant free will. This is defined as the ability of someone to choose between good and evil. Just as it would be absurd to praise a robot for emptying the trash when it was programmed to do that and nothing else, so too would it be absurd to praise someone for making a good moral decision if they had no moral choice in the matter. Consider possible worlds:

World 1: God gives people free will, and does not force them to always choose good; there is evil and suffering.

World 2: God does not give people free will, and forces them to always choose good; there is no evil and suffering.

World 3: God gives people free will, and forces them to always choose good; there is no evil and suffering.

World 4: God gives people free will, and does not force them to always choose good; there is no evil and suffering.

Plantinga says that World 3 is logically impossible; even God cannot force someone to freely do something. World 2 contains only automatons, and is hence undesirable on classical theism. This leaves World 1 and 4. World 1 describes the actual world, and World 4 describes a way the world could be: everyone has free will, but they all choose to do only good. The Book of Genesis hints that God intended the world to be like this.

Flew and Mackie maintain that God could and should have created World 3; Plantinga maintains that this is logically impossible.

Divine omnipotence and the Free Will Defense

Some scholars maintain that "omnipotence" should mean the ability to anything, even logically impossible things. But there are many things God cannot do: lie, cheat, be ignorant, etc. This supports the thesis there are things even an omnipotent being cannot do.

Objection: Free Will and Natural Evil

The free will defense does not say anything about natural evil (earthquakes, mosquitoes, etc), and so some think it is incomplete. Plantinga suggests that invisible demons could be responsible for natural evil and hence, are abusing their free will in  exactly the same way humans do. This response seems implausible, but to rebut a logical inconsistency all that needs to be shown is a possible way it can be reconciled, even if that is not the real reason for it or even if its plausible at all.

Was Plantinga's victory too easy?

J.L. Mackie, mostly associated with mid-20th Century atheism, concedes that Plantinga did away with the logical problem of evil. But he says that it is still unsatisfactory because Plantinga only deals with the possible and not the plausible; Plantinga does not give any evidence for his claim. However, most philosopher's agree that the blame lies with Mackie in making too strong a claim for the problem of evil and as a result most debate has turned to the evidential problem of evil instead.

Evaluation of the Free Will Defense

If a situation is claimed to be impossible, all that needs to be shown is a way for that situation to be possible. The possible situation does not need to be actual or even plausible, as long as it is conceivable then it is logically possible and hence the strong claim that the situation is impossible is refuted. The free will defense seems to fulfill this need and hence, the logical problem of evil is generally considered to be refuted.

Other Solutions to the Logical Problem of Evil

John Hick suggests that evil and suffering are part of a plan of "soul making," that is, shaping humans into the best that they can be. Evolution was a part of this process. This response is a theodicy instead of a defense: an attempt to provide an actual reason for evil rather than just a possible reason, as Plantinga does.

Eleonore Stump suggests that evil is what brings people closer to God; that without evil, constant joy and happiness would make people in love with worldly things rather than God.

Problems with the Free Will Defense

Heaven is usually conceived of as a place where it is impossible for people to do wrong. But if this is the ultimate desirable reward, then why is free will so valuable here on earth? If people lose their free will in heaven then are they the same people they were on earth? And if they keep their free will but are forced to always do good, then doesn't this mean that World 3 (above) is actually possible after all? If so, then Flew and Mackie's objections above are correct.

Secondly, If God is incapable of doing wrong, then it seems he does not have free will and hence, is not morally praiseworthy; this conflicts with classical theism. But if he does have free will but is incapable of doing wrong, then this seems to make World 3 (above) possible after all. If so, then Flew and Mackie's objections above are correct.