The Best of All Possible Worlds: Leibniz and His Critics

The best of all possible worlds is the claim, made by Gottfried Leibniz in his 1710 Theodicy, that God, being perfectly wise, good and powerful, chose to create the best world there could be. Voltaire mocked it, and most philosophers of religion now doubt that a best world exists at all. The classical answer runs between them: God's act of creating could not be done better, though the world He made could always be bettered, as Thomas Aquinas taught four centuries earlier.

Key takeaways

• Leibniz's thesis is a claim about God's choice. A perfectly wise and good God, he argued, "cannot but have chosen the best," so this world, evil included, is the best one possible.
• Aquinas drew the distinction Leibniz missed. God cannot make anything with greater wisdom and goodness than He makes it, yet "He can always make something else better than each individual thing."
• The modern debate turned on freedom. Robert Adams argued that God may create less than the best without wronging anyone; William Rowe replied that an improvable choice would make God surpassable.
• Frame-Indexed Classical Theism (FICT) removes the ranking. Where Leibniz's optimism and Adams's account of grace both picture God choosing from a ranked list, it holds that "best world" measures the Creator's act on a creature's scale. This world is very good and ordered to redemption.

What does "the best of all possible worlds" mean?

The phrase names a thesis about creation. Leibniz coined it in his Essays of Theodicy on the Goodness of God, the Freedom of Man and the Origin of Evil (1710), and Wikipedia calls it "the central argument in Leibniz's theodicy," his attempt to answer the problem of evil.

Leibniz's best possible world is the conclusion of a short argument. In the Monadology he compressed it into one sentence: "the wisdom of God permits him to know it, his goodness causes him to choose it, and his power enables him to produce it." Three attributes of God are at work in it. He knows every world that could exist, wills the best of them, and has the power to make it. By "world" Leibniz did not mean a planet. "I call 'World' the whole succession and the whole agglomeration of all existent things," he wrote. The best world is the whole history of everything, from its first moment to its last.

The thesis turns the problem of evil inside out. The ordinary problem asks how a good God could permit so much suffering. Leibniz answers that He permitted exactly what the best world contains, and that a world with less would have been worse. If that is true, every horror in history belongs to the best whole there could be. If it is false, a second question opens: could a perfectly good God make anything less than the best?

Free — the complete book

Get Book 1: Resolving the Free Will Dilemma free when you enter your name and email.

This essay takes one piece of it. Resolving the Free Will Dilemma takes the whole dilemma from first axiom to conclusion — the complete book, free.

No spam. Unsubscribe in one click.

The Epicurean paradox and the wider question of why God allows suffering sit on either side of this one. Here the question is narrower. It is not whether God permits evil, but whether the world that contains it must be the best.

Is this the best of all possible worlds? Leibniz at his strongest

Leibniz answered yes, and his case is stronger than its reputation, because it begins from God's perfection rather than from the state of the world. In the Theodicy he writes: "Now this supreme wisdom, united to a goodness that is no less infinite, cannot but have chosen the best." The reason follows at once: "a lesser good is a kind of evil if it stands in the way of a greater good; and there would be something to correct in the actions of God if it were possible to do better." A God who could do better and did not would have a fault to answer for.

The second step comes from the principle of sufficient reason. If no world were best, none would be preferable to another, and a perfectly wise being would have no reason to choose. So "if there were not the best (optimum) among all possible worlds, God would not have produced any." That God created at all is Leibniz's evidence that a best exists. Nor did he think the choice was forced. As the Stanford Encyclopedia puts it, "God, though free, is 'morally necessitated' to choose the best," bound by wisdom as a wise person is, without compulsion.

What of evil? The world, Leibniz answers, is one connected whole: "the universe, whatever it may be, is all of one piece, like an ocean." Remove one evil and the whole changes, so that "if the smallest evil that comes to pass in the world were missing in it, it would no longer be this world." He grants that "one may imagine possible worlds without sin and without unhappiness," but insists "these same worlds again would be very inferior to ours in goodness." He concedes, "I cannot show you this in detail," and asks us to judge from the effect. Then he quotes the Easter liturgy's "O felix culpa," and later calls the Fall "a fortunate sin, because it had been expiated with immense benefit by the incarnation of the Son of God."

The modern form of the question is harder. Most philosophers of religion now hold that there is an "infinite hierarchy of increasingly better worlds," which, the Stanford Encyclopedia on divine freedom reports, "appears to be the dominant view in contemporary analytic philosophy of religion." William Rowe and others turned this into "the problem of no best world." If every world God could make has a better rival, then whatever God makes, His act could have been better, and a being whose act could be better is not unsurpassable. At full strength the question is a dilemma: either there is a best world and this is it, horrors included, or there is no best world and no maximally perfect God.

Did Voltaire refute Leibniz?

He made the thesis unforgettable, which is not the same thing. The Lisbon earthquake of 1 November 1755 turned Leibniz's optimism into a target, and in 1759 Voltaire published Candide. Its tutor, Pangloss, "proved admirably that there is no effect without a cause, and that, in this best of all possible worlds, the Baron's castle was the most magnificent of castles." His teaching ends in a correction of ordinary optimism: "they who assert that all is well have said a foolish thing, they should have said all is for the best." Wikipedia notes that Pangloss parodies Maupertuis as well as Leibniz. Through war, shipwreck and the gallows, he keeps saying it. The novella ends with a resolve: "I know also," said Candide, "that we must cultivate our garden."

Satire shows that the thesis is almost impossible to say beside a grave. It does not show that the thesis is false. Pangloss is a caricature, for Leibniz never held that each evil is good, only that the whole containing it is best. The serious objection came later, from inside philosophy of religion, and it was aimed at the word "best" itself.

Must God create the best? The standard answers at their strongest

Christian thinkers have mostly answered no, in four main ways, and each answer protects something true.

The classical answer. Augustine set the pattern. In the Enchiridion he wrote that God "would never permit the existence of anything evil among His works, if He were not so omnipotent and good that He can bring good even out of evil." Augustine claims that God brings good out of evil, not that the world is the best possible. Thomas Aquinas asked Leibniz's question four centuries before Leibniz, "Whether God can do better than what He does?", and answered with a distinction. "Absolutely speaking," he wrote, "God can make something else better than each thing made by Him." Yet if "better" describes the manner of the making, "God cannot make anything better than He makes it, because He cannot make it from greater wisdom and goodness."

The making cannot be bettered; the thing made can. The present universe, Aquinas adds, cannot be bettered on its own terms, since altering one part would spoil the order of the whole, "as if one string were stretched more than it ought to be, the melody of the harp would be destroyed," and yet "God could make other things, or add something to the present creation; and then there would be another and a better universe." The view, kept by Thomism, protects God's freedom and His aseity, and it fits the classical account of omnipotence. Its strain is the one Leibniz pressed: if no world is best, what reason did God have for this one?

Grace instead of maximizing. In 1972 Robert Adams argued that God could make a world less than the best without wronging anyone, provided that, in the Stanford Encyclopedia's summary, no creature in it has "a life which is so miserable on the whole that it would have been better for that creature if it had never existed." Choosing such a world could even show grace, "a disposition to love which is not dependent on the merit of the person loved." This frees God from a duty to maximize. Rowe replied that Adams shows "that God's moral perfection imposes no moral obligation" to make the best, "But this isn't quite the same thing as showing that God's perfect goodness does not render it necessary" that He make the best. Wronging no one may still be the lesser act.

The happy fault. Alvin Plantinga gave a Christian reason for thinking our world ranks among the best. As Agustín Echavarría summarizes him in TheoLogica, "Plantinga claims that God's permission of evil is a function of the good-making features of the world, among which we find the Incarnation and Atonement." He called the view supralapsarian, Echavarría explains, because on it "God's decision to save some man through Incarnation is in some way logically prior to His decision to allow original sin." Worlds with a Redeemer outrank worlds without one, and a Redeemer presupposes sin. The strength is that Christ becomes the reason for creation. The strain is that sin seems to become a means God chose.

The best God could get. Open theism holds that God gave creatures libertarian freedom and took a real risk with it. The world is not the best possible, but the best God could achieve with free agents whose choices He does not settle. That clears God of authoring evil and gives human choices real weight in how history goes. It also leaves Him unable to guarantee the end. The believer is promised a God who will do His best with the world rather than one who has settled how it ends.

If you want these accounts weighed chapter by chapter, Volume One is free to download.

How Frame-Indexed Classical Theism answers the best-world question

Frame-Indexed Classical Theism (FICT) is a research program in analytic theology that resolves classical theological dilemmas by diagnosing them as artifacts of a single error: one term carried across two frames as though it meant the same thing in each. Here the term is "best." Leibniz's optimism, Adams's account of grace and Plantinga's happy fault share one picture, in which God surveys the worlds He could make, ranks them and takes one. They differ only on whether He must take the top. Open theism keeps the ranking and lowers God's control of the outcome.

"Best" is a ranking word. It compares finite options on one scale, from outside them, and selects. That is how a creature chooses, because a creature stands before alternatives it did not make. Carried into God's own frame, the word pictures God at a moment of deliberation. Volume Four, Resolving Immutability & Interaction, refuses that picture: "God does not choose among worlds fixed in advance by wills He must accommodate." His purposes are secured "not by screening the creature's outputs beforehand but by encompassing every one of them." The survey picture belongs to Molinism, which needs it. The classical account of God's sovereignty does not.

Once the frames are told apart, Leibniz and Aquinas each keep what they saw. Leibniz was right that God's act could not be bettered, since it proceeds from infinite wisdom and goodness. That is Aquinas's adverb. Aquinas was right that the thing made could always be bettered, since on the creature's scale of finite goods there is no top, and no finite world exhausts an infinite goodness. That is his noun. Rowe's argument needs the two to be one, so that an improvable product implies an improvable act. Indexed, the act is unsurpassable and the product is unbounded, and the two claims do not conflict. Scripture's verdict on creation is likewise no ranking: "God saw all that he had made, and it was very good" (Genesis 1:31, NIV).

Why, then, this world? Volume Two claims consistency, not optimality. God permits finite evil "for morally sufficient, overarching redemptive purposes," and that reason is "ordered to this end: to bring finite beings, through redemption and sanctification, to a state in which they freely maintain righteousness by works in unbroken communion with God." Volume One explains the redeemable creation: God "ordered a creation whose eventual fall would be redeemable rather than irretrievable," so that grace and mercy could be shown to the undeserving. Christ "was chosen before the creation of the world, but was revealed in these last times for your sake" (1 Peter 1:20, NIV). "You intended to harm me, but God intended it for good to accomplish what is now being done, the saving of many lives" (Genesis 50:20, NIV).

Does this dodge the question of why God made this world? It does not. The reason is given: the world is ordered to redemption and to creatures who freely maintain righteousness in communion with God. What is refused is the ranking, the further claim that this reason makes our world the top entry on a list of every world God could make. Paul states the promise that the reason rests on: "And we know that in all things God works for the good of those who love him, who have been called according to his purpose" (Romans 8:28, NIV).

Whether God had to create at all belongs to modal collapse, and why He created belongs to why God created. God's goodness itself is the subject of omnibenevolence, and the statement of the position sets it out in one place.

Volume Two, Resolving the Problem of Evil, carries this answer through natural evil, judgment and heaven.

Five attacks on the best-world answer

We have run thousands of simulated attacks against Frame-Indexed Classical Theism, drawn from academic and peer-reviewed sources. Five of them bear directly on the best of all possible worlds, from the side of Leibniz's optimism, of Plantinga's felix culpa and of Robert John Russell's eschatological theodicy.

Is "God couldn't do better" enough for each person?

Here's one of those attacks and how the Many Beings framework would address it.

Marilyn McCord Adams pressed the sharpest objection to best-world reasoning in her 2008 critique of Plantinga in Faith and Philosophy. She distinguished God's goodness in producing an excellent cosmos from His goodness to each person He creates. A world can be excellent on the whole and still use someone as a means, as when one person's ruin serves the repentance of others. That God could not have done better, she argued, does not show that He loved that person. One must also show that no one's life is so bad that, as Jesus said of Judas, "It would be better for him if he had not been born" (Matthew 26:24, NIV).

The reply begins with what her standard requires. It measures God's absolute goodness by the finite life of each sufferer, so that God is good to a person only if every horror is defeated within that person's own life. That carries a creature's measure of good into God's frame. It also proves too much, since applied consistently it entails that all are saved, which Volume Two rejects. The book therefore declines the standard rather than failing it, and states its own claim exactly: God's permission of finite evil is consistent with His goodness, not a promise that each horror is defeated for its bearer within this life. The claim is not cold. God's patience is aimed at each person, "not wanting anyone to perish, but everyone to come to repentance" (2 Peter 3:9, NIV), and present troubles are "achieving for us an eternal glory that far outweighs them all" (2 Corinthians 4:17, NIV).

Does the happy fault make sin God's chosen means?

Here's one of those attacks and how the Many Beings framework would address it.

Plantinga's felix culpa argument is the modern Christian best-world argument. If the best worlds contain Incarnation and Atonement, and atonement requires sin, then God permits sin for the sake of redemption. Plantinga called this supralapsarian, because the decision to redeem comes first in God's intention and the permission of sin serves it. The attack turns this on Volume Two. That book lists redemption in Christ among God's reasons for permitting evil, yet it also grounds the permission in the fact that finite evil does not oblige God to eradicate it at once. One order makes redemption the end; the other makes it a response. Which is it?

The reply is that the two answer different questions, and Volume Two states both in one premise. God is under no obligation to eradicate every finite evil immediately, and He "permissively allows the temporary operation of Finite Evil" for "morally sufficient, overarching redemptive purposes." The first removes a constraint, and the second gives the reason. A removed constraint is not an order of intention. God decrees in one simple act, which the book never divides into earlier and later decrees, so its claim of consistency needs no ruling between supralapsarian and infralapsarian. Paul holds the same line: "where sin increased, grace increased all the more" (Romans 5:20, NIV), yet to "Shall we go on sinning so that grace may increase?" he answers, "By no means!" (Romans 6:1-2, NIV). Grace overflows sin without making sin desirable. The same distinction governs God's permissive will: He permits what He does not endorse.

Doesn't God still pick one world from a list?

Here's one of those attacks and how the Many Beings framework would address it.

This attack comes from inside the series. Volume One, Resolving the Free Will Dilemma, says that God, rather than creating only an order of angels, created humanity so that its fall would be redeemable, "precisely so that the full range of the divine nature could be manifested" through redemption centered on the Cross. That sounds like comparing two possible creations by their results and choosing one. Yet Volume Four denies that God surveys options and picks. Does the series take back with one hand what it gives with the other?

It does not, because the two passages address different things, and each says so where it stands. Volume One places its claim "At the trajectory level," in the structure of a creation, and insists that "At the particular level, Adam genuinely could have stood." It compares two modes of creaturely freedom, not two catalogues of what particular creatures would choose. Volume Four's denial answers a narrower question: must God first learn what each free creature would do in every circumstance, then choose the world whose outputs suit Him? It rejects "a moment of deliberation at which God surveys the options and picks." Ordering a structure toward an end is what any wise maker does. Screening free choices in advance is what the classical God has no need to do, and it is the screening that the best-world picture requires.

If heaven has free creatures who never sin, why not create heaven first?

Here's one of those attacks and how the Many Beings framework would address it.

James Sennett posed the dilemma in Faith and Philosophy in 1999. If the redeemed in heaven are free and never sin, God could seemingly have created free, sinless creatures from the start, and this world's evil was unnecessary. Sennett's own answer, as Timothy Pawl and Kevin Timpe report it, is "that God couldn't have actualized a world in which humans are free from the start and also are incapable of sin in virtue of being determined never to sin." The attack argues that Volume Two lacks that answer, because it describes heaven's stability as a gift of conditions (a perfected nature, God's presence, sustaining grace), and gifts could be given at creation.

The reply sits beside the attacked line. The glorified "continue to make genuinely finite righteous choices, but those choices now express a confirmed orientation rather than decide an open one." A confirmed orientation is one that was decided, and a creature placed directly into glory would have nothing to confirm. Volume Two finds the same structure in angels, who can "be decisively confirmed in righteousness or decisively inverted in rebellion." It also names the good that requires the route: God grants this span of moral agency from "His longing for genuine relationships, wherein humans may freely choose to align with Him through faith in Christ." The post on free will in heaven takes up the rest.

If the new creation has no natural evil, why not start there?

Here's one of those attacks and how the Many Beings framework would address it.

Robert John Russell, writing on entropy and evil, compares two orders of creation: the present cosmos, where the same processes that give life also bring decay and death, and the new creation, where they no longer do. If embodied, free persons will live in the new creation without earthquakes or decay, then natural evil is not needed for such persons, and the present world cannot be the best that could be.

The reply is that the two orders differ in kind, not degree. Volume Two treats the resurrection body as "a creaturely change in nature," a new condition that is not "more of the same." Freedom differs with it: present agency decides an open orientation, while glorified agency expresses a settled one. The same volume holds that "A world governed by consistent natural laws is a necessary precondition for the existence of embodied, free moral agents," agents still deciding who they will be. The new creation shows that confirmed agency needs no natural evil. It says nothing about what the deciding life requires. When that life is finished, the promise stands: "He will wipe every tear from their eyes. There will be no more death or mourning or crying or pain, for the old order of things has passed away." He who was seated on the throne said, "I am making everything new!" (Revelation 21:4-5, NIV).

The Many Beings series works through objections like these across seven volumes.

Frequently asked questions

Who said "the best of all possible worlds"?

Gottfried Leibniz coined the phrase in his 1710 Theodicy, as the conclusion of his argument that a perfectly wise and good God must create the best world He can. Voltaire made it famous in Candide (1759) by putting it in the mouth of the tutor Pangloss, as satire.

Is there a best possible world?

Probably not. Most philosophers of religion now hold that for any world there is a better one, as there is no largest number. Aquinas held the same: God "can always make something else better than each individual thing." What cannot be bettered is the wisdom and goodness with which God makes. William Rowe turned the endless series into an argument against God's existence, but it works only if God's act is measured on the same scale as the world.

Could God have created a world without evil?

Leibniz granted that such worlds can be imagined but judged them "very inferior to ours in goodness," because the world is one connected whole and removing a single evil would make a different world. The classical answer is that God permits finite evil for reasons consistent with His goodness and ordered to redemption. Whether God created evil is a separate question about its origin.

Does God have to create the best world?

Leibniz said yes. Robert Adams argued that God wrongs no one by creating less than the best, and Frame-Indexed Classical Theism adds that "best" ranks the world on a creature's scale. God's act cannot be bettered, and the world He made is very good.

Is the Leibniz best possible world argument the same as Pangloss's optimism?

No. Pangloss treats each event as for the best, down to his claim that "the nose has been formed to bear spectacles." Leibniz argued that the whole world is the best possible even though it contains sin and suffering, which he admitted are evils. Voltaire's caricature is funnier than Leibniz's argument, and less accurate.

Where to go from here

Candide's last line is often read as a shrug. Read again, it comes close to the truth. We are not asked to rank the worlds God could have made. We were given this one, very good, wounded, promised its renewal, and asked to tend it. Even Pangloss, in the novella's last pages, recalls that man was first put in Eden "that he might cultivate it." The best of all possible worlds was never ours to pick out. The question we can answer is whether the God who made this one is good.

Volume One is yours free, and the series applies the same method to evil, time and changelessness.