The free-will defense is an argument commonly attributed to Alvin Plantinga, who developed it as a response to the logical problem of evil. However, in developing this argument Plantinga unwittingly ended up reinventing/rediscovering the Molinist doctrine of middle knowledge, so the key ideas of the argument are not entirely novel, and people have certainly come up with similar defenses independently more than once.

The Wikipedia article includes a summary of Plantinga's argument:

A world containing creatures who are significantly free (and freely perform more good than evil actions) is more valuable, all else being equal, than a world containing no free creatures at all. Now God can create free creatures, but He can't cause or determine them to do only what is right. For if He does so, then they aren't significantly free after all; they do not do what is right freely. To create creatures capable of moral good, therefore, He must create creatures capable of moral evil; and He can't give these creatures the freedom to perform evil and at the same time prevent them from doing so. As it turned out, sadly enough, some of the free creatures God created went wrong in the exercise of their freedom; this is the source of moral evil. The fact that free creatures sometimes go wrong, however, counts neither against God's omnipotence nor against His goodness; for He could have forestalled the occurrence of moral evil only by removing the possibility of moral good.

I have the following objection to this argument:

  • If human beings were created in the image of God and have free will, then it follows that God also has free will.
  • Likewise, if human beings, in virtue of having free will, are capable of moral evil, then God, in virtue of having free will, must also be capable of moral evil.
  • However, if God is omnibenevolent, He is not capable of moral evil.
  • So it looks like we have a contradiction between the last two points.

Said in another way, if God can have free will and be incapable of moral evil at the same time, then why would God create human beings that have free will and yet are not incapable of moral evil at the same time?

In other words, God is a counterexample to the claim that free will necessarily entails being vulnerable to moral evil, since God has free will and yet is not vulnerable to it, and so one wonders why God would create free creatures that are not immune to moral evil, just like He is.

How do proponents of Plantinga's free will defense against the problem of evil resolve this conundrum?

In case anyone feels the knee-jerk compulsion to downvote this question as off-topic, I invite them to downvote these questions as well, posted by high-rep users:

Based on some of the comments received, I will try to write a more formal and rigorous version of the objection:


  • P1: God is omnibenevolent
  • P2: God is omnibenevolent => God is not capable of moral evil
  • P3: God has free will
  • P4: God has free will => God is capable of moral evil


  • D1: God is not capable of moral evil (from P1 & P2)
  • D2: God is capable of moral evil (from P3 & P4)
  • Contradiction between D1 & D2 (=><=)

Defense of the premises

  • P1 (God is omnibenevolent) is probably uncontested. Pretty much everyone concedes this as an axiom in the definition of God.
  • P2 (God is omnibenevolent => God is not capable of moral evil) should be uncontroversial as well. God cannot do evil. It's impossible/unfeasible for Him. It just won't happen.
  • P3 (God has free will) is based on the intuition that if humans (and angels) have free will, it would be very strange for God not to have free will as well. One could reject this premise and claim that, perhaps, God is a deterministic being who created free creatures. Sure, one could hold such a view, but it would be a very novel (and strange) one, wouldn't it?
  • P4 (God has free will => God is capable of moral evil) is based on the same intuition used by the free-will defense against the problem of evil. If evil is explained as an undesired price of having creatures with free will (which God was willing to pay because of how valuable free will is), then what the defense is basically saying is that free will => capable of moral evil. So P4 is just a particular application of that rule to God, if we concede that God has free will.

If anyone is interested in reading defenses of Plantinga's argument from different perspectives, you can do so by paying a visit to this question.


3 Answers 3


The Free Will Defense against the Problem of Evil is relatively unique compared to the more common responses. Most responses fall into the general category of "God does not have omnibenevolence", and some are that "God does not have Omnipotence", and a few reject the concept of agency -- that God would act to implement His will. This defense holds that the basic Problem of Evil is correctly stated, but that there is a previously unrealized property of Free Will that makes it a Good that counteracts all evil.

This argument, if one spells out its premises, has the following form:

  • The Problem of Evil is a valid argument, God would create the best of all possible worlds to achieve His objectives, and there is great evil in this world
  • However, we are mistaken about the nature of Omnibenevolence, and Free Will
  • Free Will is so important, that its existence vastly overbalances all the evil in this world
  • IF we have Free Will THEN we will do great evil
  • And this evil we do of our own free will is the sole Evil in the world
  • Therefore this Good of Free Will makes this the best of all possible worlds

There are a variety of ways to critique this set of assumptions.

Your critique is entirely valid, although it is only one of several problems with this "free will" argument. God, as you note, IS free. Yet does not do evil, at least per the omni-benevolence theology. You are offering a falsifying test case, demonstrating that Plantinga's rationale is missing a few key elements.

What the main flaw you are poking at here is, is that what one does is not just based on freedom, but also upon inclinations. One can be entirely free to do massive amounts of harm, but also have no inclination to DO harm. Such a person would, entirely freely, do no harm to anyone. So -- what inclinations do we humans have, that we do harm often? Clearly we are such that we have a significant propensity to commit harm, yet an omni-Creator COULD have created us without such a harm-leaning propensity.

This refutes decisively one key point in this argument: "IF we have Free Will THEN we will do great evil".

So -- if we could have had just as much freedom as we do now, AND be more loving, caring, and helpful than we tend to be, THEN God as creator would be responsible for all the evil we do because of the inclinations that we have, and God is once again the responsible party for the evil we do.

There are several other objections to this argument, which address other flaws in it. One of those is that this argument, which presumes that the "good" of free will trumps all the significant evil that it presumes free will requires to therefore happen in the world, does not address that our human free will is very limited. We have limited knowledge, power, imagination, and ability to control our own inclinations, AND we are highly constrained by the choices of others. Humans could be MORE free if we had more knowledge, more power over the world around us, stronger imaginations and judgment about how to achieve our goals, stronger wills to overcome and/or mold our inclinations, and more autonomy relative to the constraints other humans put on us. IF free will were SO morally valuable as this argument presumes, that even our limited free will is worth more moral weight than all the evils of this world, then any step to increase our free will should similarly be such a great good as to outvalue any other considerations, and any creator God should therefore have maximized our Good of Free Will. Any creator who left us with such a limited portion of free will, would have created an immensely morally deficient world, compared to one in which we could have had so much greater freedom, leaving this world once again far from the morally best of all possible worlds.

Additionally for a third objection, most of the evil in the world is not created by humans, it is intrinsic in the Darwinian equation. Life multiplies until it is living on the ragged edge of survival due to scarcity. AND life can only survive by destroying other life. Our universe is structured to cause scarcity, competition, and massive quantities of violent death. This is evil built into the universe at its foundations, and no degree of human freedom or lack of freedom has any effect on its intrinsic moral imperfection.

So -- your critique is valid, plus there are at least two more that are also highly effective in refuting Plantinga's argument.

What you will likely find in response to your critique, is not any defense of Plantinga's Free Will argument, but a resort to the more common arguments against The Problem of Evil -- IE special pleading that God's "good" is different from and unknowable from our "good" (God is not omnibenevolent), or that God is unable for some reason to optimize the world (God is not omnipotent, or not an agent).

The reason the "Problem of Evil" remains a "problem" is that none of the rationalizations that have been trotted out to try to deflect from it, stand up to scrutiny.

  • our "good" No such thing. People disagree on what is good.
    – user76284
    Commented Jun 14, 2023 at 17:34
  • 2
    @user76284 Disagreement over the good can be dealt with as we do with other noisy data. There is no need to deny realism because our data is noisy. See the answer I posted to one of your questions: philosophy.stackexchange.com/questions/44977/…
    – Dcleve
    Commented Jun 14, 2023 at 18:18
  • 1
    @Dcleve The disagreement over "good" isn't always noise. Especially in this case, it is the disagreement of relativism vs absolutism: "our good" differs from "God's good" (or "THE good" or what-have-you) only to a relativist
    – qxn
    Commented Jun 14, 2023 at 19:41
  • 1
    @Dcleve There's no agreement on what "our good" is, so that phrase should be replaced with "what I consider good" or the like.
    – user76284
    Commented Jun 15, 2023 at 22:52
  • 1
    @Dcleve I agree with your statement about disagreement, and don't dispute that the word can be used with little controversy in certain contexts. But this is not one of those contexts. What constitutes "our good" in a philosophical context is far more contentious than what constitutes "a sandwich" in a quotidian context.
    – user76284
    Commented Jun 16, 2023 at 4:58

One could fairly trivially say that being capable of evil (but not committing evil) is compatible with the definition of omnibenevolence.

Capable of evil or does evil?

However, there is a stronger point here if you replace the capability to do evil with actually doing evil. If humans do evil because of free will, the same should apply to God. Plantinga's argument only mentions free will, so this should serve as a rebuttal.


To address the above, one could add that we are also imperfect or impure or we have flawed characters, while God is perfect. But this would undermine Plantinga's argument: if we're imperfect, this is already sufficient to explain why we might do evil, and you therefore don't need to invoke free will.

Whether it would make for a good argument to invoke imperfection is another question entirely. It's certainly much less compelling on the surface level: free will is something we value (putting aside whether free will conceptually makes sense), so it's easy to accept this as a reason for why people do bad things. But it's less convincing to say it was necessary for us to be created imperfectly (or with the specific imperfections that would lead to us doing evil).

God is always good?

Some have addressed the issue by arguing that anything God does is, by definition, good, therefore anything he has the will to do would necessarily be good. But then you have to ask whether it would be good to e.g. throw babies off a building if God commands it.

  • There are a few people who have gone to the extreme of saying that, yes, this would be good if God commands it. But then "good" would mean little more than "what God is and does", which doesn't really tell us anything.

    One could similarly define "good" as "what NotThatGuy is and does". And then anything I do would by definition be good. But disregards questions of whether my actions harm others, whether I only care about myself, etc. (i.e. what factors into how "good" is typically defined).

    It may not strictly speaking be "wrong" to define "good" in this way, but then we could come up with a new term for the original definition of "good", i.e. whether someone tries to prevent harm to others, whether they're considerate, caring, respectful, kind, selfless, etc. We can call that "good2". Then we'd be back to the question of whether God is "good2" (but he doesn't seem to be, based on the evidence available).

  • But most would say God wouldn't command people to throw babies off a building. If this is because God is good, then this is circular: God wouldn't not do good because God is good (the obvious response: prove that God is good). It also ignores quite a few things God commanded (or personally did) in the Bible, like drowning basically everyone, including babies (the flood), killing firstborns (Egypt) and raining fire on cities to wipe out everyone, including babies (Sodom and Gomorrah), to name a few things. I suppose it depends how literally you take what's written in the Bible (but it's written in there either way).

Another problem: why create people this way?

I think the far bigger problem with the argument is that if God knows what will happen, and creates us exactly as he wishes, why would he choose to create beings that will choose to commit atrocious evils, only to then punish those beings for acting consistent with the character that God created them with?

Also, why would he allow them to go through with their evil acts, at the expense of the suffering of others? If I have 2 children, I may allow them to play freely, even if it comes with the risk that one would harm the other. But if I see one is about to about to cause the other egregious harm, and I have the means and opportunity to stop them, and I don't, I hope we'd all agree that I'd be an absolutely horrible parent. Yet, this is exactly what God supposedly does every day.

What about the problem of suffering?

I prefer the closely related problem of suffering (which is often discussed as part of the problem of evil). I don't consider the free will argument to be at all compelling, but it also can't be used at all to explain why children get cancer, why there are insects that lay eggs behind people's eyes, why there are worms that burrow underneath people's skin and need to be extracted in a painful process across months, why natural disasters ravage settlements, etc.

Some say this ultimately serves some greater good (that may just be beyond our understanding), but this is just an assertion that's entirely unfalsifiable and that's supported by little more than some verses in a few-thousand-year-old book. Looking at all the immense suffering in the world, it seems really unconvincing to say all of that somehow serves a greater good. On the flip side, someone else could similarly say all the good in the world ultimately serves some greater evil (if you suffer all the time, you kind of get used to it and it stops being all that impactful - moments of joy are necessary to maximise how bad the suffering is). You're not judging God's character according to what God does and what God created, but rather you're trying to make reality fit into what you think God's character is. It's possible to fit pretty much any set of facts into any character traits. But that doesn't mean it's the most reasonable conclusion to draw from the facts. Imagine applying this same logic to a person: John is a good person. Yes, he lies, cheats, steals and assaults at every opportunity, but each of those actions ultimately serve some greater good that we may just not be able to see (John is also really smart, you see - much smarter than any of us). Would you find that convincing? I doubt it.

Others would say the "fall of man" brought evil into the world, but this is just an assertion that's entirely unfalsifiable and that's supported by little more than some verses in a few-thousand-year-old book. It's also questionable to punish all of humanity (and possibly the animal kingdom too) for the sins of Adam and Eve. The existence of a literal Adam and Eve is also in conflict with evolution. Some people mean a metaphorical fall, which sidesteps these problems, while introducing the problem of in what way the metaphorical applies to the literal real world.


I'm not sure that Plantinga had an absolutely consistent point of view in this connection. For example, he separately proclaimed a "victorious" instance of the ontological-argument scheme, one based on maximal-being theology mixed with possible-worlds talk. Per that argument, Plantinga described God as necessarily maximally good, or maximally good in all possible worlds. So there'd be no possible world in which God sinned enough to detract from His moral glory, say (whether God could sin at all and retain His glory, I don't know what Plantinga thought to say; Calvinists are sometimes unusually open to the abstract possibility of God willing evil, or willing something it could be reasonable to call "evil").

Or consider Davis and Franks[18]:

Of course, given the very nature of God as “perfectly good and holy, all-powerful and all-knowing” (7), it follows that “any world in which God exists is enormously more valuable than any world in which he does not exist” (7). ... For ease of reference, then, let J refer to Jesus’ [haecceity]. The question at once arises: does J suffer from [transworld depravity (TWD)]? Naturally, being depraved is the very last thing one wants to attribute to Jesus—at least if we’re thinking of him (as Christians do) as the one who atones for human sin and wrongdoing. ... then we don’t have it that Jesus’ essence is even possibly transworld depraved, in which case J, from all appearances, appears to be happily sheltered from the atonement-negating effects of TWD. Unfortunately, however, things are not as they appear. It turns out that there are a few complications, arising from particular features of Plantinga’s modal metaphysics.

They go on, at first glance (and I'm not fit to second-guess them, here), to show that through all this rigamarole, we end up with a divine being that "suffers from" TWD, or which is logically incapable of sinning, or whatever manner of vexing thing; there doesn't look to be a stable outcome to these pictures of things.

Also consider (again, Plantinga's! but reported in the SEP) contemplation of the "noetic effects of sin":

On the one hand, [sin] carries with it a sort of blindness, a sort of imperceptiveness, dullness, stupidity. This is a cognitive limitation that first of all prevents its victim from proper knowledge of God and his beauty, glory, and love; it also prevents him from seeing what is worth loving and what worth hating, what should be sought and what eschewed. It therefore comprises both knowledge of fact and knowledge of value. (Plantinga 2000: 207)

Presumably, even if Plantinga's God is not absolutely omniscient simpliciter, He is self-omniscient (knows Himself through and through). Or, then, God would seem to be fixed knowing whatever is needed to be known, in order for sin to be concretely impossible to choose (although again, per my first paragraph, there might be some abstract sense to be had for, "God might, just might, not be impeccable," here).

Incidentally, the general family of free-will defenses seems easy to undermine if one simply asks, "If free will in the relevant sense means the ability to make choices, and if one can have a choice between neutral-and-good apart from good-and-evil, why couldn't God create a being with the free will to either do nothing at all, only neutral things, or only good things, but nothing evil?" There seem to be at least two flavors of "mere permissibility" and two flavors of good, i.e. obligation and supererogation. The forbidden is a fifth thing besides those; so why would it not be enough for God to grant us the ability to choose from only the other four options? I imagine Plantinga might have a reply to this encoded into the labyrinth of transworld-depravity-as-a-concept, but per the above-quoted assessment of TWD as applied to Plantingan theodicy, I wouldn't want to hold my breath while waiting to find out that he does have a specific, stable reply, though.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .