Renown rationalist G. W. Leibniz (in)famously answered the problem of evil by insisting that our world is the best of all possible worlds. I’ve just finished my book on Leibniz’s “optimism” for Cambridge University Press. As shocking as his claim is, my book argues that Leibniz’s reasoning is virtually impossible for classical theists to avoid.
I’m offering my readers something rather unconventional: a preview of the volume. I have posted every chapter here. All subscribers have access to a preview, and paid subscribers can see the full manuscript.
To all my subscribers, thank you for subscribing. To my paid subscribers, thank you for your support. And to any visitors, please consider subscribing and supporting my work!
Chapter 2
Perhaps the Only Possible World
…[A]ccording to an ancient but by no means forgotten tradition, the idea of freedom is said to be entirely consistent with the idea of system, and every philosophy which makes claim to unity and completeness is said to end in denying freedom.
— F. W. J. Schelling1
Leibniz’s theory of the best raises a thorny question: Does God have free choice? Recall that the theory is based on several necessary truths. God exists. God knows the best. God does the best. The natural consequence seems to be that God has no choice but to create our world. Anything else would be incompatible with his nature. But in what sense, then, is God free?
The question of divine freedom naturally spills over into creaturely freedom. Leibniz argues that every event in our world is ultimately traced to God willing the best. This includes “free” events. But if God selects which world is real, and this selection includes which of my choices are real, then how are my choices my own and not God’s? In what sense could I have done otherwise? In what sense am I free?
The traditional reading of Leibniz is rather pessimistic about such questions. He’s read, nearly without exception, as a determinist.2 The only question is whether he is a hard or soft determinist. The reasons are many, but before going further, we should define some terms.3
“Determinism” and Other Terms
Determinism is the view that for every event, there are conditions such that nothing else could’ve happened. Applied to freedom, determinism is the belief that there are conditions that precede our choices that are sufficient to determine those choices. Our choices are products of these prior conditions and could not be otherwise.
There are various types of determinism, and each has a different reason for thinking nothing could be otherwise.4 But regardless of the reason, all determinists agree that the choices we make are inevitable, given the state of affairs in which they occur.
Incompatibilism is the view that free choice and determinism are incompatible. So, if determinism is true, then free choice is false. If free choice is true, then determinism is false. The dichotomy creates a fork in the road between hard determinism and libertarianism, both of which accept the dichotomy. The former affirms determinism and rejects freedom, while the latter affirms freedom and rejects determinism.
The telltale sign of libertarianism is what has been called the “Principle of Alternative Possibilities” (PAP). This principle identifies the central libertarian assumption, namely, a free act is one that entails alternative possibilities, or could have been otherwise, or was avoidable. Determinism is incompatible with PAP, since the defining feature of determinism is that nothing could be otherwise. So, the libertarian insists that one must choose, freedom or determinism. The libertarian chooses the former. Our choices reflect PAP.
Here we arrive at the position known as compatibilism. Before saying what compatibilism is, allow me to say what it is not. Compatibilism is not the view that libertarian freedom and the will of God are somehow compatible. Unfortunately, many in theological circles hear “compatibilism” and presume it affirms a theological paradox, where we have libertarian freedom but this freedom is compatible with God’s foreordinations. This is not what the term means.
Compatibilism is the view that freedom and determinism are compatible. This likely sounds strange, since determinism denies that things could be otherwise and the commonsense view of freedom is that our choices could be otherwise. This, however, is where compatibilism works its magic. Compatibilism denies that free choice requires PAP, and thus redefines “free choice” to mean “any choice made in accord with one’s desires without coercion.” So long as these conditions are met, it doesn’t matter whether the choice could be otherwise.
A thought experiment might help. Let’s say that physical determinism is true. Everything is one unbroken chain of matter in motion, wholly determined by the laws of physics, like an infinitely long string of dominoes. Within this state of affairs lives Jane, who faces a choice between p and q. Granting physical determinism, Jane is nothing more than matter in motion; her desires and choices are just physical events within this chain of material causes. And nothing in the chain could be otherwise. If Jane chooses p over q, is her choice free?
According to the libertarian, the answer is No. Jane has no control over her desires or choices, and thus could never choose otherwise. Compatibilism “solves” this problem by removing PAP from the definition of freedom. Since Jane desired p, and Jane chose p, and no one put a gun to Jane’s head to coerce p, Jane’s choice is free. Presto chango, free choice (redefined) and determinism are compatible.
Compatibilism is notoriously misleading because it uses language of freedom and even speaks about the possibility of alternative choices. For example, “Jane could do otherwise if she desired.” But what goes unsaid is that Jane herself would need to be otherwise to desire otherwise. So, for all its talk of freedom, the compatibilist is a determinist. Period. The compatibilist simply changes the meaning of “free” to exclude contrary choice, and then says that determinism and freedom are compatible. It would be like saying “square” and “circle” are compatible if we redefine “square” to mean “a geometric shape with a flowing circumference on which all points are an equal distance from a common center.” Most miss this sleight of hand.
Now, returning to Leibniz, virtually all read him as a determinist. Some read him as a hard determinist, others as a compatibilist (or soft determinist), but nearly all agree that he is a determinist of some kind.5 As we will see, once this presumption is granted, it becomes difficult, if not impossible, to avoid the conclusion that all things are absolutely necessary, despite what Leibniz might say to the contrary.
Psychological Determinism in Leibniz
For Leibniz, nothing evades PSR, not even psychological events. So, Leibniz insists, no choice is made without reason. Of course, not every choice is reasonable, but when someone chooses p over ~p, there’s a reason.
Bertrand Russell argues that psychological determinism naturally follows.6 He notes that Leibniz affirms not only PSR but the law of excluded middle. This law states that every proposition is either true or false; there is no “middle.” So, statements about future choices, like “Bob will choose to eat breakfast tomorrow,” are either true or false before the choosing. So, the psychological event is predetermined. And PSR demands a sufficient reason for this predetermination. Things don’t just happen. The simplest explanation is there’s something about Bob that determines his choice in that circumstance.7
Leibniz tries to avoid this conclusion by distinguishing inclining influences from necessitating influences. The conditions under which we choose only ever incline our will, never determine it, necessitating its choice.8 But few find this distinction successful. Why? Leroy E. Loemker says the distinction fails because Leibniz denies the very possibility of equilibrium. The will is never indifferent to its options. And whatever its dominant inclination, that’s what it chooses. Without equilibrium, any inclining influence is a determining influence, just as any imbalance in a balance scale tips it. If choice is always the product of a dominant inclination, then inclining influences are determining influences.9
A second escape Leibniz tries builds on his observation about a posteriori truths. Recall that there is no analytic necessity in pairing “Caesar” with “crossed the Rubicon.” So we can negate “crossed the Rubicon” of “Caesar” without contradiction. The alternative is possible. Hence, this truth (or choice) is contingent. Nicholas Rescher is quick to point out, however, that this only rescues Leibniz from logical necessity, the sort of necessity that follows from terms, like in geometry. But Leibniz cannot possibly mean that an event might not have followed from its causes. They are, after all, the sufficient reason for the event.10
Russell finds nothing in Leibniz to break this chain of necessity. Like Loemker, he points to equilibrium: “[Leibniz] rejected entirely the liberty of indifference … and even held this to be self-contradictory. For it is necessary that every event should have a cause….”11 What, then, causes a psychological event? Leibniz’s answer is “spontaneity,” the subject’s inner power of self-movement (G 6:122-3).
This emphasis on an inner cause should sound familiar. Recall that compatibilism emphasizes, not contrary choice, but uncoerced choice. If the cause of the choosing is my own impulse, then the choice is “without constraint.”12 So what is the inner reason for a choice? Russell finds only one answer in Leibniz: The idiosyncratic nature.13 And with this, we stand at the precipice of determinism. If a subject’s nature is the sufficient reason for his choice, and this determining nature is had by the subject prior to and without his choosing, then the subject’s self-movement reflects, not contrary choice, but his innate disposition. His “freedom” is determined by his nature.
Might the addition of reason evade this conclusion? After all, Leibniz distinguishes bare spontaneity from spontaneity paired with reason, which is why animals are not free but humans are.14 Does this stave off determinism? Russell thinks not.
He points out that, for Leibniz, rational volition is determined by knowledge of the good, this being its target.15 But here emerges an inevitability. The will is determined by desire, “and if the good means what is desired, then volition would be necessarily determined by the good.”16 The point is obvious in Leibniz’s theology. God’s volition is determined by the best.17 In the case of creatures, however, the will is finite and ignorant. So a creature’s will is determined, not by the good, but by the perceived good. Yet, the connection between judgment and action is no less necessary. Whatever the creature perceives to be best he chooses. The only difference between us and God is the accuracy of our aim. 18
Russell, therefore, sees all Leibniz’s talk of contingency as nothing but a smokescreen: “All these remarks are discreditable subterfuges to conceal the fact that all sin, for Leibniz, is original sin, the inherent finitude of any created monad, the confusedness of its perceptions of the good, whence it is led, in honest and unavoidable delusion, to pursue the worse in place of the better.”19
Jack Davidson offers the same conclusion from a different angle. He points out that Leibniz sees God as the “paradigm of freedom, and we are only free in so far as we are like him.”20 But God necessarily does the best. His “freedom” is not contrary choice. And Davidson argues that God is not unique in this regard. All free beings act in accord with their greatest inclination and their perception of the best.21 What sets God apart is his perfect perception of the best.22 The hypothetical necessity that he does the best makes him more free, not less. Such a view requires compatibilism.23 For the apex of “freedom” excludes alternative possibilities.
In the end, Leibniz’s commitment to PSR demands that every choice is made for a reason. For God, this reason is his perfect knowledge of the best. For creatures, the reason is our imperfect knowledge of the best. But in either case, the choice reflects the chooser’s dominant inclination and perception of the best. Our choices may not be logically necessary, as four-sided is to square, but they are nonetheless inevitable products of our nature.
Superessentialism in Leibniz
I said above that, for Leibniz, our choices may not be necessary, only inevitable. But perhaps even this is too charitable.
In his correspondence with Antoine Arnauld, Leibniz says we can talk about Adam doing otherwise in Eden (not eating from the tree), but if Adam had chosen otherwise, he would not be “our Adam.”24 The claim is surprising. But some reflection on incompossibility and concomitance makes it less so. Persons are not bare abstractions. We are part of a complex web of interconnected realities, many of which cannot be negated without negating us. God knows possible persons, not independent of all worlds, but as members of possible worlds. And this web of concomitants includes a web of choices and actions. Hence, our Adam is a person who belongs to our world, and God’s knowledge of him and our world includes his sin in Eden. To posit Adam doing otherwise is to posit a different world and a different Adam.
The point is evident in Leibniz’s 1686 essay, “On Necessary and Contingent Truths” (NCT) (C 16-14), where he suggests that every truth is analytic from God’s perspective. For us, some truths are a posteriori, since we are finite and ignorant. But God, being omniscient, knows all things apart from any investigation. If we knew Caesar the way God knows him, we would see in Caesar everything true of him, including crossing the Rubicon.25 From this Louis Couturat argues that Russell’s reading of Leibniz is too charitable, since it allows for compatibilism — the deed could be otherwise, though the choice is inevitable. But if every deed is contained in the very idea of the subject, so that if we saw the subject as God does we would see his deeds, in what sense could the deed be otherwise?26
David Blumenfeld explains how this raises the specter of “superessentialism,” the idea that “every property that an individual has (save existence) is an essential part of his nature.”27 Leibniz can talk about the semantic coherence of a phrase like “Adam did not eat of the tree in Eden,” but if doing so makes Adam someone other than himself, it would appear that sinning in Eden is essential to Adam — as essential to him as four-sided is to square.