Friday, May 05, 2006

The Stasis of Utilitarianism in Kantian Ethics

Kant and Mill represent to us two opposites of ethical thought. The discovery of any common ground between them which might form the basis of a relationship may appear a hopeless task. Kant says that a good will is not good because of what it performs or effects, but simply by virtue of the volition, that is, it is “good in itself” and considered by itself is to be esteemed much higher than all that can be brought about by any inclination. Mill says in Utilitarianism that the “creed which accepts as the foundations of morals Utility, or the Greatest Happiness Principle, holds that actions are right in proportion as they tend to promote happiness, wrong as they tend to produce the reverse of happiness.”

We have here a conflict between the subjective and the objective—between the individual and his environment. With Kant the moral ideal is regarded as the product of pure reason. With Mill and utilitarianism it is the result of an accumulated experience of pleasure and pain. Yet before accepting this opposition as absolute, and rejecting on or the other system, we must examine the problems which were before the minds of Kant and Mill, that we may be sure they are giving contradictory answers to the same question. It may be that they are not arguing over the same thing, and that the stasis of debate is unclear. If so, we may take comfort in the idea that human reason is not as divided against itself as it sometimes appears at first glance.

The common, and apparently essential, characteristic which it would seem possible to discover in all forms of Utilitarianism is that moral conduct is estimated by the effects of action, and that the effect which is the moral criterion is the “greatest happiness for the greatest number.” By “greatest happiness” is meant the greatest excess of pleasure over pain, considered in reference either to all sentient beings or to all mankind. Pleasure is to be considered abstractly and purely quantitatively, as if one could hypothetically measure the sum of all hedons and pleasurable experiences. Thus, no question of individuality is considered. One man’s hedons are not only quantitatively different from that of another.

Most importantly, motives have nothing to do in deciding the worth of an action, except if they are useful in the production of pleasure. As Mill puts it, “The morality of an action depends entirely upon the intention—that is, upon what the agents wills to do. But the motive, that is, the feeling which makes him will so to do, when it makes no difference in the act makes none in the morality; though it makes a great difference in our moral estimation of the agent, especially if it indicates a good or bad habitual disposition—a bent of character from which hurtful or form which useful actions are likely to arise.”

In this case the motive is also included in the moral judgment, but does not form the principle of that judgment. It does not determine the moral character of the effects, but is determined by them. Thus all departments of moral life are subordinated to feeling: that is morally good which produces pleasurable feeling; that is morally bad which does not. The great merit claimed for this system of morals is its great clearness and practicality. Every one is supposed to know what pleasure is, and most people know what will tend to produce it. If they do not, the common code of morals, which is claimed to be the result of the accumulated experiences of the race, furnishes a generally correct guide. This brief description of the utilitarian position attempted to show its wholly external character. The effects of conduct must be the only objects of moral judgment, unless we fall into confusion and mysticism, losing all standards of valuation except the individual moral sense, or an arbitrary code of revealed morality. Motives, so far as they are internal, a supposed free will or autonomous self, cannot be subjected to observation and experiment, and hence must be ignored in considering the principles of moral judgment.

Kant, on the other hand, has always been connected with a system the exact opposite of the one just described. The idea is that the only good thing in the world is a Good Will. Here we have the direct contrary of the Utilitarian position that the will has no value except in relation to the effects produced by it. Kant recognizes that face that his doctrine of the absolute value of the good will must seem wholly paradoxical to the adherents of the happiness theory, yet he emphasizes it as the only basis for a universally binding moral law. If the desire for happiness were to be admitted by Kant as the principle of determination of the will, no necessary law could be formulated; since pleasurable consciousness is the result of experience, no examination would give rise to any sort of necessary law. In other words, the principle would be heteronomy instead of autonomy, which alone gives a basis for distinctly moral judgment. To allow a material principle such as pleasure to determine the will would be to place the Ego under natural laws in which no freedom is possible—and if there is no freedom, then for Kant, no morality is possible.

The moral law, then, must be purely formal and given by the subject. The good will must consist in willing the good out of pure respect for law. This is where the contrast is most clearly drawn by Utilitarians between them and Kant. Their criticism is that from such an empty statement of morality no practical duties can be deduced. What practical help will it give to apply Kant’s formulation of the moral law to any given perplexity? “Act from that maxim which you can at the same time will to be a universal law of nature”—this is the first statement of the rule.
In the case of lying, this could give us no help, since if we wished to lie it would be easy to suppose our maxim as a universal law. We might will that all men should steal, or all men should lie, and Kant’s rule would make such actions moral, since we could imagine them and will them to be a universal law. Only if we consider the consequences of the act, say his critics, can we tell whether it is moral or not. We know by experience that lying, stealing, murder, are not productive of more pain than pleasure, hence we call them morally bad, and not because they cannot be made universal laws.

Kant, however, is brilliantly guarded against such objections by his making the test of fitness logical. When we lie, we do so under the supposition that the law is for me to tell the truth; and that, consequently, men will believe that we are doing so. At the same time that we lie, we would not will that lying should be universal, because our own lie would not be believed. Plus the very distinction between truth and falsehood would vanish. Such a law would be self-destructive. So in stealing, we assume that the law of property will generally hold; for otherwise our own act would not be stealing—there would be no property-rights to violate.

It’s not the unpleasant consequences of such a society which leads to the impossibility of willing these laws, but the self-contradictory nature of the laws themselves. This point is not as important as the Utilitarians make it out to be, however. The point of real value in Kant’s thought, and the only point which it is the object of this paper to defend is not his exposition of how the moral law is to be applied to particular cases, but the determination of the principle involved in our judgment of worth. For Mill, if motives are to be judged good or bad, it is only on account of their effects. Kant, on the other hand, places all moral value on the motive itself—respect for law—obligation. Duty he defines in the Metaphysics of Morals as “the necessity of acting from respect for law.” Even actions in accordance with law are not moral, but merely legal, since they may be done from motives other than that of respect for law. Kant’s doctrine is not, “Thou shalt because it is good,” but “Thou shalt, therefore it is good.” The form of the law determines the end, and not the end the form of the law.

Moreover, nothing but the will must determine the will, and in order to find a universal and necessary determinant for all rational beings, only the bare form of will and not its material content must be taken. The law must not be externally imposed; else it loses it binding character and allows us the options of disobedience to its commands under condition of accepting the penalties attached. The governing principle must come from within—it must be the reason issuing its commands to itself. Thus not only must the idea of law be the principle, but it must be the law of autonomy—that of the self-legislating Ego. But, as we have seen, this law is deprived of all content in that it can be merely the universal form of law. To determine our moral judgment by means of an external end would be the dethroning of the legislative self. The natural criticism then is, of what use is this discovery of a law which commands the fulfillment of no end? What use is a will which wills nothing but itself? As before stated, this is the main criticism of Kantian ethics—its purely formal character and its opposition to at least a partial and popular doctrine of happiness as a good and an end of action.

Kant, however, by no means considers his system in such a debilitating way. Kant is as well aware as his critics are that one cannot will a maxim in general and at the same time will no actions in particular. He knows that there must be an object in every act of the will. The purpose of his insistence on the purely formal element in morality is to emphasize the fact that moral law must be universal and necessary—must command obedience categorically and of itself. In this, Utilitarians are right in considering his system as utterly opposed to their own, but the opposition does not consist in the fact that he denies the necessity of an end for every action. It is his aim to change the stasis of debate, i.e. the point of emphasis, and supplement their partial system by one which embraces all the phenomena of the moral consciousness.

Kant says, “So act as to treat humanity, whether in thine own person or in that of any other, in every case as an end withal, never as a means only.” Here, however, we find an apparently out of place element introduced in this conception of an end. Yet this second statement of the law is what follows from the implicit contents of the first. To say that man is to act from a principle which he can will to be a universal law—to make him the supreme law-giver to himself, is to place him above the reach of all law other than his own, or that which he freely makes his own. If, then, as a rational being he cannot be subject to anything without him, he is an end unto himself—an absolute end so far as rational. All men, moreover must also be regarded as ends, and so treated, i.e., the ends of each individual must be made the ends of the particular subject.

Kant is opposed to utility, not as an end of conduct, but as a motive to conduct. Utility places all morality in the outward manifestation. Kant would seek a law which shall command the transition from the individual to the universal—which shall enable the subject to transcend the limits of his private interests, and find his true self in the proliferation of the common good. This can only be done by a universal element in the individual. A system, on the contrary, which makes the standard of morality an end external to the subject, or else and end limited to the subject, cannot bind the subject at all, or compel him to pass beyond his own interests. If the greatest happiness of the greatest number is the sole principle of morality, what power is there which can compel man to seek it? What hold has it over the individual when conflicting with his own interest? The end must have some attractive power in itself to draw men’s wills toward its realization. But, according to Mill, this end can only appeal to men through pleasure, which, as individual, can only be estimated by the individual. If, then, someone does not feel the desire to promote the greatest happiness of anyone beyond himself, he is at liberty to act as he pleases, since there exists no common principle to which appeal may be made. His desires will from his only rule of action. If his own happiness and that of others cannot be shown to be identical, which is possible only under presuppositions destructive of utilitarianism, there seems no way of advancing from the on to the many.

Kant’s moral theory takes us out of the relative and introduces us to the sphere of true being and the Good Will. Instead of making experience the guide of life, and constructing the ideal out of experience, he sets out a theory which is universal and absolute. The content of his moral theory gives rise to experience, but the necessity of the realization is found entirely in the rational self.

Monday, May 01, 2006

Is God's Perfection Analytically Necessary?

Perhaps the most persistently misunderstood aspect of Leibniz’s thought is his principle of contingence. The theory is deeply rooted in Leibniz subject-predicate logic, and it is therefore not surprising, given the complexity of this philosophy, that he would be vulgarly misunderstood—most famously as Dr. Pangloss from Voltaire’s Candide. What I have tried to do in this paper is to clarify Leibniz’ theory of contingent existence, to defend it against a criticism about moral perfection, and ultimately to show there is a legitimate place for this theory in Leibniz’ philosophy.

Leibniz theory of truth rests on a procedure that he terms “analysis.” This is the elimination of defined, complex ideas by making analytic use of their definitions. What Leibniz calls the “Principle of Sufficient Reason” is a principle asserting that, if a proposition is true, then it is possible to show that its predicate is contained in its subject by means of an analysis or demonstration which may or may not proceed infinitely, (and in the case that it requires infinite analysis, God alone can carry out the analysis fully.) Another important principle Leibniz calls the “Principle of Contradiction” is to the effect that if the analysis of a proposition shows its predicate to be contained in its subject after a finite number of steps, then the proposition is true. Such finitely analytic true propositions Leibniz says are “necessary truths” or “truths of reason.” On the other hand true but infinitely analytic propositions are “contingent truths” or “truths of fact” (M. § 31-33). It is clear that the Principle of Contradiction is the principle of necessary truths.

So, according to the Principle of Sufficient Reason every true proposition is finitely or infinitely analytic in the mind of God. However, according to the Principle of Contradiction every finitely analytic proposition is true. The converse of the Principle of Contradiction is every true proposition is finitely analytic, which is not the same as the Principle of Sufficient Reason. Neither of these principles guarantees the truth of infinitely analytic propositions, which is to say that contingent truths are by virtue of this not necessarily true, although accepting contingent truths as truths is not on this account irrational. And since the Principle of Sufficient Reason concerns truths, there must be a further principle where a supply of infinitely analytic truths can be of some use.
First, it is essential to remark that, with the exception of contingent propositions whose subject is God, contingent truths concern contingently existing things. For, on Leibniz’ account, a contingently true proposition must have a subject. Now, this subject cannot be an “abstract object,” since truths concerning these are necessary. And, propositions about God aside, the only remaining entities on the Leibnizian account are the contingently existing things. God’s mind, on this account, is where the possible-worlds semantics can be examined more closely. Of the infinitely-many possible worlds (contained with infinitely-many possible substances) God selects one, the best, and “actualizes” it, to use Alvin Plantinga’s Neo-Leibnizian phraseology.

Every possible substance is a member of some possible world, and its complete notion involves its entire history in the development of that possible universe (D.M. § 9). Now thanks to the “pre-established harmony,” there corresponds to every state in the development of a possible substance a state of every other possible substance of its possible world: a correspondence capable of varying degrees of closeness of agreement between its members (M. § 80-82). Thus within a possible world every substance “represents” every other substance more or less “distinctly.” In other words, it perceives the other substance with a greater or lesser degree of “clarity” or “confusion” in the plenum of interconnected monads. In this way, at each stage of its development every possible substance “perceives” or “mirrors” its entire universe, and moreover it does so more or less clearly according as the mean value of the degree of clarity of its perception of individual substances varies. Leibniz calls the degree of clarity with which at a given state a possible substance mirrors its universe its amount of perfection for that state (M. § 54). Now what Leibniz terms the “amount of perfection” of a possible substance is a measure of its amount of perfections of a possible substance is a measure of its amount of perfection for all states. So, every possible universe also has an amount of perfection: the sum of the total amounts of perfections possible in the substances belonging to it.
The principle by which God selects among the possible worlds the best of them—one with the greatest maximization of “order” and “variety”—call this the Best-Possible-Worlds Principle (Theodicy § 120, 124; M. § 58). This principle is a formulation of the thesis that in His decision of creation God acted in the best possible way: according to it the actual world is that one among the possible worlds which an infinite process of comparison showed it to be the best.

The Best-of-Possible-Worlds Principle specifies that in nature some quantity is at a maximum or a minimum. It requires mathematical techniques similar to those found in calculus. This principle enables us to understand what Leibniz means concerning contingent truths as analytic, but requiring an infinite process for their analysis. A given proposition concerning a contingent existence is true, and its predicate is indeed contained in its subject, if the state of affairs characterized by this inclusion is such that it involves a greater amount of perfection for the world than any other possible. So it is the infinite comparison required by the Best-Possible-Worlds Principle that in infinite process is imported into the analysis of truth dealing with contingent existence.

It will be made clear that Leibniz’ Principle of Contingence is his Best-Possible-Worlds Principle. And it is in virtue of this principle that infinitely analytic propositions can be truths. Leibniz writes to Arnauld that “a contingent existent owes its existence to [the Best-Possible-Worlds Principle], which is sufficient reason for existents.” Leibniz calls the “necessity” of contingent truths moral necessity as opposed to the metaphysical necessity of necessary truths, and he states that “moral necessity stems from the choice of the best.” In Section 46 of the Monadology Leibniz speaks of “the contingent truths whose principle is that of suitability or of the choice of the best.” And he maintains that “contingent propositions have demonstrations… based on the principle of contingence or existence… on what seems best among the several equally possible alternatives.”

Now, the Principle of Sufficient Reason demands exactitude. It states that a contingent truth is susceptible of an analysis which, though infinite, converges on something. But such exactitude could equally well have been gained had God chosen the worst of all possible worlds. The Principle of Sufficient Reason requires merely that contingent truths are analytic. The Best-Possible-Worlds Principle shows how this is the case. As Leibniz repeatedly said, the Principle of Sufficient Reason leaves open to God’s choice an array of alternatives for possible actualization—of which the best possible world is the only one. Therefore, though it is true that the Principle of Sufficient Reason requires some complementary principle of exactitude, Leibniz would have been the first to deny that this must be the Best-Possible-Worlds Principle.

So far we have been occupied with the theory of contingent existences. That is, the principle that all true propositions are analytic, finitely or infinitely (Principle of Sufficient Reason); that all finitely analytic propositions are true (Principle of Contradiction); and that all infinitely analytic propositions—and thus all propositions whose infinite analysis converges on some characteristic of the best of all possible worlds—are true (Best-Possible Worlds Principle). At this point we are faced with an objection which would rule contingence out from the system of Leibniz once and for all. The objection is that the contingence of God’s goodness and the necessity of His existence is a contradiction within Leibniz’ system, and in which the Problem of Evil seems to be at hand. On the Leibnizian account, either God’s goodness is contingent or it is necessary. Should Leibniz have held that both God’s existence and goodness are necessary, and succumb to a kind of Spinozistic determinism? (And what might “necessary goodness” refer to on the Natura Naturans thesis?) This thesis is quite attractive, although accepting it would collapse nearly all of Leibniz into Spinoza. In formulating a “religiously adequate” account of God as the necessary being, Leibniz was not prepared to affirm both and say that “God’s goodness is a necessary predicate of His existence as a necessary being.” Nor was he prepared to say that “God’s existence is contingent and therefore His goodness is too.”

Leibniz maintains the existence of God as the ens necessarium (“necessary being”) and invokes a modal argument to show that the ens necessarium exists provided only that its existence is possible. If the proposition “God exists” is possible on this account, then ipso facto it is rational to believe it is true, says Plantinga (Necessary Being, Free Will Defense). It remains for Leibniz to demonstrate that “God exists” is possible.

Leibniz’ own argument is one from causality, found in the Leibniz-Arnauld correspondence: if the ens necessarium is not possible, then no existence is possible. If the ens necessarium is possible, then He exists. If we say that ens necessarium does not exist, then it would follow that nothing exists. But something does exist. Hence the ens necessarium exists. In other words, God is a causally necessary condition of the existence of anything else, but whereas His own existence has no necessary conditions. Now the absence of a necessary condition of the existence of anything else is a sufficient condition of the nonexistence of that thing. And if a being has no causally necessary conditions, then its nonexistence has no causally sufficient conditions. And hence if God does exist, His going out of existence could have no causally sufficient conditions and is therefore causally impossible. If God has no necessary conditions, then it is analytic that His “going out of existence,” if it occurred, would be an uncaused event, since there are no causally sufficient conditions for that occurrence. Likewise, His “coming into existence” is causally impossible, since it is analytically true that God is not dependent upon anything else. He has no cause, and therefore His coming into existence would be an event which would have no causally sufficient conditions. So, if God does exist, He cannot cease to exist; nor could He have begun to exist.

But God’s goodness is not included in the above argument. Let me now put it in. Since God exists necessarily—His existence being contained in His essence—it follows that God has the highest possible degree of perfection. Perfection in this sense is His metaphysical perfection, containing “as much reality as is possible,” which Leibniz made clear in the Monadology and to Arnauld. But God is purported to be perfect not only in the metaphysical sense, but perfect in the moral sense as well. While Leibniz is certain of God’s moral perfection, he does not maintain it is necessarily true. He maintains that while His existence is necessary, His goodness as a creator, i.e. moral perfection, is contingent and the result of His free choice. And hence, being the good creator as He is, He freely chooses to be morally perfect.

We must cope now with the question of the relation between God’s necessary metaphysical perfection and His contingent moral perfection. In order to do this we must once more call to mind Leibniz Principle of Sufficient Reason. As we have already seen, this asserts that every true proposition can be show to be analytic by a (possibly infinite) process of analysis. Using this principle we can clarify the logical relation between the two types of divine perfection. Divine moral perfection has a sufficient reason, and this in turn another sufficient reason ad infinitum. But this sequence of sufficient reason converges on God’s metaphysical perfection, on His existence. Or putting it another way, we can say that God’s moral perfection is indeed a logical consequence of His metaphysical perfection, but a consequence which cannot be proven logically within a finite number of steps. In this way, as Leibniz insists, the proposition asserting God’s moral perfection is contingent. God is only morally perfect by free choice, not necessitation.

Thus it is precisely the infinite regress which naturally invokes a reductio ad absurdum response to God’s goodness being contingent. But Leibniz implicitly maintains that infinite processes are not ipso facto erroneous, since a kind of mathematical convergence is possible. To assert convergence is to say there is a limit to the infinite series, and may be itself unknown. Leibniz, after all, invented calculus independently of Newton. And it is his notation which is has been in general use since.

The philosophy of Leibniz rests on the three fundamental principles discussed in this paper. It is the Best-Possible-Worlds Principle, not the Principle of Sufficient Reason, which constitutes Leibniz’ contingency theory and his moral perfection theodicy. It is the Principle of Sufficient Reason which allows for the distinction between God’s metaphysical perfection and moral perfection, and enables Leibniz to maintain both the contingency of God’s goodness and the necessity of His existence.

Truth in Leibnizian Philosophy

The most persistently misunderstood aspect of Leibniz’s thought is his principle of contingence. The theory is deeply rooted in Leibniz subject-predicate logic, and it is therefore not surprising, given the complexity of this philosophy, that he would be vulgarly misunderstood—most famously as Dr. Pangloss from Voltaire’s Candide.

What I have tried to do is to clarify Leibniz’ theory of contingent existence, to defend it against a criticism about moral perfection, and ultimately to show there is a legitimate place for this theory in Leibniz’ philosophy.

Leibniz theory of truth rests on a procedure that he terms “analysis.” This is the elimination of defined, complex ideas by making analytic use of their definitions. What Leibniz calls the “Principle of Sufficient Reason” is a principle asserting that, if a proposition is true, then it is possible to show that its predicate is contained in its subject by means of an analysis or demonstration which may or may not proceed infinitely, (and in the case that it requires infinite analysis, God alone can carry out the analysis fully.)

Another important principle Leibniz calls the “Principle of Contradiction” is to the effect that if the analysis of a proposition shows its predicate to be contained in its subject after a finite number of steps, then the proposition is true. Such finitely analytic true propositions Leibniz says are “necessary truths” or “truths of reason.” On the other hand true but infinitely analytic propositions are “contingent truths” or “truths of fact” (M. § 31-33). It is clear that the Principle of Contradiction is the principle of necessary truths.

So, according to the Principle of Sufficient Reason every true proposition is finitely or infinitely analytic in the mind of God. However, according to the Principle of Contradiction every finitely analytic proposition is true. The converse of the Principle of Contradiction is every true proposition is finitely analytic, which is not the same as the Principle of Sufficient Reason. Neither of these principles guarantees the truth of infinitely analytic propositions, which is to say that contingent truths are by virtue of this not necessarily true, although accepting contingent truths as truths is not on this account irrational. And since the Principle of Sufficient Reason concerns truths, there must be a further principle where a supply of infinitely analytic truths can be of some use.

First, it is essential to remark that, with the exception of contingent propositions whose subject is God, contingent truths concern contingently existing things. For, on Leibniz’ account, a contingently true proposition must have a subject. Now, this subject cannot be an “abstract object,” since truths concerning these are necessary. And, propositions about God aside, the only remaining entities on the Leibnizian account are the contingently existing things. God’s mind, on this account, is where the possible-worlds semantics can be examined more closely. Of the infinitely-many possible worlds (contained with infinitely-many possible substances) God selects one, the best, and “actualizes” it, to use Alvin Plantinga’s Neo-Leibnizian phraseology.

Every possible substance is a member of some possible world, and its complete notion involves its entire history in the development of that possible universe (D.M. § 9). Now thanks to the “pre-established harmony,” there corresponds to every state in the development of a possible substance a state of every other possible substance of its possible world: a correspondence capable of varying degrees of closeness of agreement between its members (M. § 80-82). Thus within a possible world every substance “represents” every other substance more or less “distinctly.”

In other words, it perceives the other substance with a greater or lesser degree of “clarity” or “confusion” in the plenum of interconnected monads. In this way, at each stage of its development every possible substance “perceives” or “mirrors” its entire universe, and moreover it does so more or less clearly according as the mean value of the degree of clarity of its perception of individual substances varies. Leibniz calls the degree of clarity with which at a given state a possible substance mirrors its universe its amount of perfection for that state (M. § 54). Now what Leibniz terms the “amount of perfection” of a possible substance is a measure of its amount of perfections of a possible substance is a measure of its amount of perfection for all states. So, every possible universe also has an amount of perfection: the sum of the total amounts of perfections possible in the substances belonging to it.

The principle by which God selects among the possible worlds the best of them—one with the greatest maximization of “order” and “variety”—call this the Best-Possible-Worlds Principle (Theodicy § 120, 124; M. § 58). This principle is a formulation of the thesis that in His decision of creation God acted in the best possible way: according to it the actual world is that one among the possible worlds which an infinite process of comparison showed it to be the best.

The Best-of-Possible-Worlds Principle specifies that in nature some quantity is at a maximum or a minimum. It requires mathematical techniques similar to those found in calculus. This principle enables us to understand what Leibniz means concerning contingent truths as analytic, but requiring an infinite process for their analysis. A given proposition concerning a contingent existence is true, and its predicate is indeed contained in its subject, if the state of affairs characterized by this inclusion is such that it involves a greater amount of perfection for the world than any other possible. So it is the infinite comparison required by the Best-Possible-Worlds Principle that in infinite process is imported into the analysis of truth dealing with contingent existence.

It will be made clear that Leibniz’ Principle of Contingence is his Best-Possible-Worlds Principle. And it is in virtue of this principle that infinitely analytic propositions can be truths. Leibniz writes to Arnauld that “a contingent existent owes its existence to [the Best-Possible-Worlds Principle], which is sufficient reason for existents.” Leibniz calls the “necessity” of contingent truths moral necessity as opposed to the metaphysical necessity of necessary truths, and he states that “moral necessity stems from the choice of the best.” In Section 46 of the Monadology Leibniz speaks of “the contingent truths whose principle is that of suitability or of the choice of the best.” And he maintains that “contingent propositions have demonstrations… based on the principle of contingence or existence… on what seems best among the several equally possible alternatives.”

Now, the Principle of Sufficient Reason demands exactitude. It states that a contingent truth is susceptible of an analysis which, though infinite, converges on something. But such exactitude could equally well have been gained had God chosen the worst of all possible worlds. The Principle of Sufficient Reason requires merely that contingent truths are analytic. The Best-Possible-Worlds Principle shows how this is the case. As Leibniz repeatedly said, the Principle of Sufficient Reason leaves open to God’s choice an array of alternatives for possible actualization—of which the best possible world is the only one. Therefore, though it is true that the Principle of Sufficient Reason requires some complementary principle of exactitude, Leibniz would have been the first to deny that this must be the Best-Possible-Worlds Principle.