Saturday, June 28, 2014

Book Review: God, Freedom, and Evil by Alvin Plantinga

In God, Freedom, and Evil Alvin Plantinga (AP) attempts to rebut the logical problem of evil [i], which posits that the following two propositions [i] are inconsistent:

(1) God (an omnipotent and perfectly good being) exists
(2) Evil exists

Where is the Inconsistency? 
AP spends the first section of the book attempting to demonstrate an inconsistency between the two premises. He argues that (1) and (2) are neither explicitly nor formally [ii] contradictory, and (following J. L. Mackie) decides that the most promising course for the atheologian [iii] is that the propositions are implicitly contradictory. A set of premises is implicitly contradictory if one or more of the terms violates a logically necessary truth. What the atheologian is looking for, then, in pressing the logical problem of evil is a necessary truth which, when added to (1) and (2) above, yields a contradiction.

The Inconsistency Stated 
Eventually AP settles on the following formulation:
(1) God (an omnipotent and perfectly good being) exists.
(2) God, being omnipotent, could create any world he wanted.
(3) God, being all good, would prefer a world with no evil.
(4) Evil exists.

Conclusion: Therefore, God does not exist.

To put it more simply, God could and would create a world with no evil. But the world has evil. So, there must be no God.

God and Evil Have Not Been Shown to Be Inconsistent
It is important to understand what is being claimed here. It is not that evil is somehow disconfirming of God’s existence in that we are less justified in believing in God. The claim is much more ambitious than that. The claim is that, given the existence of evil, it is impossible that God exist. The argument, if successful, is devastating. On the other hand, rebutting the argument requires showing only that it is possible that God and evil coexist. And that is what AP aims to show. [iv]

Recall that for a set of premises to be implicitly inconsistent, they must violate a logically necessary truth. The purportedly logically necessary truths added were premises (2) and (3). All AP has to show, therefore, is that these added premises are not necessarily true, therefore rebutting the argument. AP addresses (3) briefly, but spends the most of his time on (2). In addressing (3), AP notes that some goods entail evils. Consider, for example, the situation in which a person endures pain with patience and gratitude towards God. Such a good state of affairs ‘contains’ the fact that evil exists. [v] It seems, then, at least possible (which is all that a defense requires) that (3) is false and thus is not logically necessary.

In arguing that (2) is not logically necessary, AP notes two things. First, some atheists have lumped in (2) with Voltaire’s sneering at the Leibnizian thesis that this is the ‘best of all possible worlds’. Surely, Voltaire judged, the world we see is not the best world there is. AP argues that Leibniz made a mistake, which he memorably coins Leibniz’s Lapse. The basic idea here is that, like numbers, there may simply be no logical maxima. Put differently, any world we imagine might be made better by adding more people, or more planets, or….etc. etc. So, just as there is no highest number, there may be no best possible world.

Second, and most central to AP’s defense, is the idea that—granted a certain view of human freedom—it may not lay within God’s power to create just any world he wishes. To do AP’s defense justice, we have to introduce a handful of terms: libertarian freedom, possible worlds, feasible worlds [vi], and transworld depravity.

Turning first to libertarian freedom, AP argues that libertarian freedom can be cashed out as ‘the power of contrary choice’. AP notes that moral good (agents doing the morally right thing) seems to imply libertarian freedom—the ability to do good or to do evil. And this surely seems logically possible. [vii, viii] Next, consider what a possible world is.[ix] For our purposes, a possible world might be defined as a total, internally consistent description of how reality might be. [x] Marrying the concepts of libertarian freedom to possible worlds, AP discusses a possible world containing a fellow named Curley who is being offered a bribe. Given their contradictory natures, one and only one of the following two propositions is true:

Proposition #1 : Curley takes the bribe.


Proposition #2: Curley rejects the bribe.

Let’s call the possible world containing Proposition #1 PW1, and the possible world containing Proposition #2 PW2. Now, both PW1 and PW2 are possible worlds - they contain no contradictions. So God should be able to create either of the two possible worlds. [xi] But remember—Curley has the power of contrary choice. It is thus up to Curley which proposition is true, and therefore up to Curley which of the two possible worlds God can create.

Before proceeding further, let’s clarify what is going on here. In constructing the logical problem of evil, the atheologian is implicitly arguing that God cannot do the logically impossible. If the atheologian is not arguing that, then how on earth is the logical problem of evil supposed to get off the ground? What AP is pressing here is the argument that it is logically impossible to make someone freely do something. And this leads us to our next term: feasible worlds.

Return to the Curley example. Let’s say that Curley (unfortunately) chooses to take the bribe. What this means is that, although PW1 and PW2 are both possible worlds, only PW1 is a feasible world. [xii] That is, although it is logically possible (there are no logical contradictions in supposing that PW2 could, in some sense, exist) that Curley not take the bribe, he will in fact do so.

And it may be that Curley would commit at least one morally wrong act in any possible world in which he was placed. If so, then there is no feasible world in which God can place Curley and Curley not commit evil. This property AP has dubbed transworld depravity. Further, it may be the case that not only Curley but all persons (actual or possible) suffer from transworld depravity. If this is the case, then it is possible that God could not create a world with no moral evil in it. [xiii]

Given the demonstration that these things are possible, AP judges that God and the existence of evil have been shown to be compatible.

Atheologian Rejoinder: God is Incompatible With the Amount of Moral Evil
However, the atheologian can regroup [xiv] and say (roughly) the following: fine, you’ve demonstrated that God and the existence of some evil are compatible. But surely God’s existence is incompatible with the amount of evil we see. That is, God could have created a world with a better balance of good over evil than this one.

Once again, naturally enough, the distinction between possible and feasible worlds comes into play. In essence, Planting argues that, yes, there are possible worlds with a better balance of moral good over moral evil than the actual world. But it is possible that there are no feasible worlds with a better balance of good over evil.

God, Freedom, and Evil is a fascinating book, and one that anyone interested in the problem of evil should read. As far as I can tell, it does successfully undermine the logical problem of evil. On the one hand, that is gratifying for the theist—this was for centuries the ‘go to’ argument for the atheologian. On the other hand, it is a testimony to how fallible human reasoning can be. Once one grasps the logic of the free will defense, it is a bit hard to see how so many generations of admittedly brilliant atheistic thinkers put such confidence in the logical problem of evil.

Apologetics 315 Book Reviewer Latter Day Inkling is a U.S.-based research psychologist for the military. He is especially interested in epistemology and natural theology.

[i] Defusing the logical problem of evil is the core (but not the whole) of the book. AP also takes a very cursory look at three theistic proofs from natural theology (the cosmological, teleological, and ontological arguments). AP judges that the cosmological and teleological arguments are failures, but that the ontological argument holds some promise. If the reader wants to think of the book summary as a scoreboard, it was Atheism (logical problem of evil) and Theism (the ontological argument) going toe-to-toe. The final score was Theism wins over Atheism, 1 to 0. (This is obviously tongue-in-cheek, and Plantinga’s vetting of the ontological argument is heavily caveated.)

[ii] As AP judges that the premises are neither explicitly nor formally contradictory, definitions of these terms can be relegated to a footnote. Explicitly contradictory would require that one of the premises is a straight forward denial of the other. That is, in addition to premises (1) and (2) above there would be a premise like (3) God does not exist and/or (4) Evil does not exist. (3) explicitly contradicts (1), and (4) explicitly contradicts (2). Formally contradictory means that, although not explicitly stated, there is a contradiction ‘hiding’ in the stated premises which can be brought out by the rules of logic. AP gives the following example: (5) All men are mortal. (6) Socrates is a man. (7) Socrates is not mortal. This set of premises does not contain an explicit contradiction, but it does contain a formal one. Namely, (5) and (6) entail (8) Socrates is mortal, which contradicts (7).

[iii] AP’s term for someone putting forward arguments for atheism.

[iv] In technical terms, AP is offering a defense as opposed to a theodicy. A defense merely offers possible reasons for God’s permitting evil and suffering, whereas a theodicy goes further and specifies what God’s reasons for permitting evil and suffering are in fact. In one sense, as AP notes, a theodicy is going beyond what is required for rebutting the logical problem of evil.

[v] Although AP for a long time eschewed a theodicy and stuck to ‘defenses’, he has more recently adopted a ‘greater good’ theodicy (akin to the ‘greater good defense’ above, wherein a good ‘contains’ an evil). In this case, the greater good is the Atonement. See his “Supralapsarianism, or O Felix Culpa” in Christian Faith and the Problem of Evil, Peter van Inwagen (Ed).

[vi] ‘Feasible’ worlds is not a term that AP uses in God, Freedom, and Evil. To the best of my knowledge the term was coined by William Lane Craig, who adapted AP’s ‘free will defense’ in discussing soteriological problems. See his 'No Other Name': A Middle Knowledge Perspective on the Exclusivity of Salvation through Christ". Faith and Philosophy 6. (1989): 172-88. The article can be found at

[vii] I would go much further and argue that libertarian freedom is not only possible but almost self-evidently actual. The discussion would take us too far afield, but I would urge the reader to take a look at the role that libertarian freedom plays not only in moral accountability but also rationality.
[viii] This type of freedom is sometimes termed ‘incompatibilism’. Libertarian freedom can thus be contrasted with compatibilist views of freedom, which are propounded both by many atheists and some theists (e.g., staunch Calvinists).

[ix] For a more technical discussion of possible worlds, see Location 388-461 of God, Freedom, and Evil (Kindle edition).

[x] The actual world is, quite obviously, a possible world.

[xi] I’m sliding over a subtle point involving the terms ‘creating’ and ‘actualizing’. It doesn’t affect anything of import here.

[xii] Another way of putting this point is to say that while all feasible worlds are possible, not all possible worlds are feasible. Giving human beings libertarian freedom may take certain possible worlds ‘off the table’. This should not give even the most orthodox theist pause, as this is simply a self-imposed constraint. If God gave human beings libertarian freedom, then he chose to accept such a constraint.

[xiii] The atheologian can, of course, press an argument based on natural evils (e.g., earthquakes) rather than moral (e.g., rape) evils. AP extends the free will defense to embrace fallen angels and points out that they might be responsible for natural disasters. AP is well aware that this will strike skeptics as absurd, but he points out that as a person mounting a defense all that is needed is that it be possible (not even plausible, much less true).

[xiv] AP presses his argument even further and attempts to show that this approach also undercuts versions of the evidential argument from evil. (Where evil is taken as evidence against, but not a logical, irrefutable disproof of, theism.) I am unsure what to make of this, as atheist philosophers do tend to agree that Plantinga’s approach has defanged the logical problem of evil—and yet these atheists also tend to propound versions of the evidential argument.


Ex N1hilo said...

Statements such as "it is possible that God could not create a world with no moral evil in it; " may score points in a philosophical joust with an atheist. But let's be clear: the deity Plantinga proposes is not the God of the Bible. YHWV has promised that He will create a world without moral evil.

Then I saw a new heaven and a new earth, for the first heaven and the first earth had passed away, and the sea was no more. And I saw the holy city, new Jerusalem, coming down out of heaven from God, prepared as a bride adorned for her husband. And I heard a loud voice from the throne saying, “Behold, the dwelling place of God is with man. He will dwell with them, and they will be his people, and God himself will be with them as their God. He will wipe away every tear from their eyes, and death shall be no more, neither shall there be mourning, nor crying, nor pain anymore, for the former things have passed away.”
(Revelation 21:1-4 ESV)

They will bring into it the glory and the honor of the nations. But nothing unclean will ever enter it, nor anyone who does what is detestable or false, but only those who are written in the Lamb's book of life.
(Revelation 21:26-27 ESV)

But according to his promise we are waiting for new heavens and a new earth in which righteousness dwells.
(2 Peter 3:13 ESV)

Given these facts, how dare we present Plantinga's lesser deity to anyone. We may win the argument, but we will have hidden the true God from them. We can't afford to divorce our apologetical and evangelistic arguments from the truths God has revealed in Scripture.

Remington B said...

>>"it may not lay within God’s power to create just any world he wishes."

This route creates a prima facie problem for Christian and Jewish theism, given verses like the following which indicates that God does get what he wants (and specifically what he wants from humans):

Daniel 4:35 all the inhabitants of the earth are accounted as nothing, and he does according to his will among the host of heaven and among the inhabitants of the earth; and none can stay his hand or say to him, “What have you done?”

Proverbs 21:1 The king’s heart is a stream of water in the hand of the Lord; he turns it wherever he will.

2 Samuel 17:14 And Absalom and all the men of Israel said, “The counsel of Hushai the Archite is better than the counsel of Ahithophel.” For the Lord had ordained to defeat the good counsel of Ahithophel, so that the Lord might bring harm upon Absalom.

Isaiah 63:17 O Lord, why do you make us wander from your ways and harden our heart, so that we fear you not? Return for the sake of your servants, the tribes of your heritage.

Jeremiah 10:23 I know, O Lord, that the way of man is not in himself, that it is not in man who walks to direct his steps.

And so forth, I could add a lot more. Now I'm not saying these verses can't be given some interpretation which make them consistent with the idea that God can't get what he wants from people, but a straightforward reading of these verses would certainly seem to indicate that God is capable of getting exactly what he wants.

Perhaps the atheist will want to point out that while some theist may be able to rely on rejecting premise 2, this route is problematic for the Christian or Jewish theist.

>>"Marrying the concepts of libertarian freedom to possible worlds, AP discusses a possible world containing a fellow named Curley who is being offered a bribe."

But since "possible" in "possible world" only means "isn't logically contradictory," then there is no clear reason I see to think anything significant is going on by "marrying libertarian freedom to possible worlds."

Consider this:

A physical determinist could grant that there is nothing *logically contradictory* in the statement "Curley rejects the bribe" but believe the proposition is infeasible because Curley has been physically determined to accept the bribe. Since the physicalist can tell the exact same story at these points, these points seem impotent for a libertarian theodicy.

I think libertarians are often fuzzy on this point because they are sneaking in the back door (perhaps unconsciously) power of contrary choice. This is what they really want to marry to possible worlds. But if they do this, they lose the feasibility point, it seems to me, because it's not mere logical possibility but some metaphysical possibility. You then end up with both being true: it's true that Curley will take the bribe in PW2 and true that he will not take the bribe in PW1. Then there is no clear basis for saying one or the other is infeasible.

>>"And it may be that Curley would commit at least one morally wrong act in any possible world in which he was placed."

In what sense may it be that Curley would commit at least one morally wrong act in any possible world? Are you just saying "For all we know, this may be true"?

>>"compatibilist views of freedom, which are propounded both by many atheists and some theists (e.g., staunch Calvinists)."

Many atheists propound libertarianism too.

>>"it may be the case that not only Curley but all persons (actual or possible) suffer from transworld depravity." [...] "If God gave human beings libertarian freedom, then he chose to accept such a constraint."

If humans get their libertarian nature from God, where do they get their transworld depravity from?

Peter Schaefer said...

Remington B-a few quick comments. (1) While some libertarians may be 'sneaking in' the power of contrary choice, that charge doesn't stick in this case. I explicitly define libertarian freedom as involving contrary choice in the review above. (2) True, while many atheists adopt a compatibilist analysis of freedom, many posit libertarian freedom. Whether or not they should do so is arguable. (See Stuart Goetz's essay 'Naturalism and Libertarian Agency' in "Naturalism: A Critical Analysis. And yes, I know atheism is not strictly identical to naturalism--but it's not an accident that there is a large overlap between the groups.) (3) Citing a few verses which show that God can and sometimes does 'harden hearts' does next to nothing to wipe away the vast swathes of Scripture which are most naturally read on the assumption of something like libertarian freedom. (4) Regarding 'for all we know this may be true', in a sense that is what I am saying. In the review I said, roughly, this is what makes G,F&E so interesting. The logical problem fails because it is vastly over-ambitious. All that is needed to keep the argument from going through is a logically possible scenario that shows the logical compatibility of God and evil. And Plantinga has done that. (5) Your paragraph beginning "A physical determinist could grant that..." Quite a lot of confusion there. First, the point at issue is not whether or not it is logically contradictory that Curley rejects the bribe. Rather, the whole point at issue was whether or not it was logically contradictory that evil acts (in this case, Curley accept the bribe) co-exist with God. So Curley's rejecting the bribe is not, strictly speaking, to the point. (6) Where does transworld depravity come from? *If* TD exists, it comes about b/c of counterfactuals of (human) freedom. Truths that are rooted in how we would freely act.

Remington B said...

Thanks for the response.

(1) And yet you speak of these possible worlds as merely denoting what is logically coherent. But if you're marrying libertarian freedom to possible worlds then PWs are more charged than bare logical possibility. And so I ask again, why is PW feasible and another isn't? Simply saying "the truth of what Curley will do" doesn't seem to work since either is true in the respective PW.

(2) It's also arguable whether or not they should accept compatibilism, if popular apologetic arguments for God and morality go through...

(3) The verses show more than that God hardens hearts, if you read the verses.

(4) I'm not sure why "for all we know this may be true" is supposed to have such traction. That approach can work for any alleged contradiction. We don't even need to bothering positing Plantinga's apologetic. Why isn't it enough to say "For all we know, God and evil are not contradictory"? Why does Plantinga think he needs to designate some story by which they aren't contradictory if he doesn't think the role of the story is to give some ring of plausibility to it? (And then why back off from plausibility later?)

(5) But apparently you think it IS worth pointing out that it's not logically contradictory that Curley rejects the bribe... for that seems to be all you're saying by positing a PW1! Obviously you want *more* than that, but for some reason you think that's a component that needs to be pointed out. Of course, even the most staunch determinist has access to this too: it's possible Curley not accept the bribe. If that sounds odd, it's only because of how *you* are cashing out possible worlds.

(6) The Christian and Jewish faiths have some story they can point to that explains why humans are depraved. There was an act of sin which brought about depravity. There is no similary grounding for transworld depravity (we'll go with TD, though that could get confused with total depravity), indeed it doesn't seem there could even be such a story. Libertarian freedom makes TD less likely, not more likely.

Peter Schaefer said...

(1) By and large atheists and theists alike think Plantinga’s FWD does defang the logical problem of evil. Why? Because the logical problem of evil was, in a sense, too narrow in scope. It didn’t take into account other constraints on worlds beyond mere logical coherence. So I fully acknowledge that possible worlds, as relevant to this discussion ‘are more charged than bare logical possibility’. That’s the point: again, atheists and theists by and large now accept that the logical problem of evil, as traditionally cashed out, failed to take into account other factors. Namely, those which might distinguish feasible worlds from possible worlds.

(2), (3), (5) & (6) are best set aside, I think. They strike me as both vague and, even if unpacked, would require too much time etc. Not a discussion for a combox.

(4) Why not simply say “For all we know, God and evil are not contradictory?” Why not provide a ‘defense in depth’? You can point out that for all we know, God and evil are not contradictory. If you are the least bit compelled to do that, you should be prepared to *defend* the ‘for all we know’. This will, presumably, involve showing that the contradictory is not explicit….but now you’re already on the path that Plantinga took. This is in fact what I tend to do when discussing these issues with skeptics. Point out that what I’m saying is not prima facie contradictory, then go further to point out possible ways that the things might cohere. I have a friend who is a young earth creationist who will discuss the kalam (which, on many formulations, accepts the old age of the earth). He can say something like “Even on your own principles, there is a problem for atheism….”etc. One could do that here with Plantinga’s approach. Of course one might be dealing with someone who wants to argue simply for the sake of arguing. But that is to be decided on a case-by-case basis.

In any case, at least I stirred up some interest with the review. You strike me as a fellow I would love to have a long conversation with.


Chent said...

Thank you for the review. But I don't think anything new here. It's only the free will defense but with a more philosophical language. Am I wrong?

Peter Schaefer said...


The book itself is almost 40 years old. So even from that perspective it's not 'new' in the sense that Lady Gaga is. But using 'possible world' semantics does make more perspicuous both what the logical problem of evil is claiming and why it fails to deliver. And that's not nothing. It's no accident that the logical problem of evil is so seldom invoked post-Plantinga FWD and was so often invoked prior.

Peter Schaefer said...

Ex Nihilo: you're confused about what a 'possible world' here means. It means a total state of affairs, past, present, and future. Heaven is in a sense a 'subset' of a possible world. A possible world includes what happens here on earth (pre-heaven) and in heaven itself. So that total state of affairs includes evil.What you are referring to as a 'world' is not what Plantinga means by possible world.

Ex N1hilo said...

Peter: If God can make a creation that excludes sin (as Scripture declares He will), then He is capable of actualizing a world with no moral evil; although He has obviously not chosen to do so.

Peter Schaefer said...

Ex N1hilo: whether or not God can actualize a world excluding sin depends on the kind of beings God chose to create.

If God chose to create beings with libertarian freedom, then you're wrong and Plantinga is right. If God chose to create beings with compatibilist freedom, then you're right and Plantinga is wrong.

Again, this does nothing to impugn God's power, any more than saying that God cannot create a world where 2+2=5 or where green weighs 3 pounds. (Note the sting in this for the atheologian: if the atheologian wants to insist that God is not bound by the laws of logic, then the whole edifice of this version of the problem of evil collapses.)

In any case, the point is (somewhat) moot, for all that is required is that Plantinga be *possibly* right (not actually right) for the logical problem of evil to fail.

I find it interesting that this confusion in discussing possible worlds arises. Ronald Nash (in an online course series, can't remember the exact location) once told a class of an exchange he had with Plantinga on this very issue. He asked Plantinga 'Well, what about heaven? Won't that be a sinless world?" Plantinga's response was exactly what my first exchange with you revolved around: as used in possible world semantics, 'world' means 'the whole show'.

Ex N1hilo said...


Thank you for clarifying the use of the term "world." I was sloppy in my use of it. But my point stands: We ought to tell folks of the God who exists as revealed in Scripture, rather than conjecture about how a god might possibly be and act.

Peter Schaefer said...

Total agreement with the first part of your last sentence re: witnessing. But even a 'maybe this, maybe that' scenario can remove barriers. It played a role in someone coming to the Lord-I've seen it work.

The Spirit will use all kinds of means.

Chent said...

Thank you, Peter, for your explanation

Post a Comment

Thanks for taking the time to comment. By posting your comment you are agreeing to the comment policy.

Blog Archive