A Defense of Leibniz’s Cosmological Argument from Contingency

God is a hot topic. Many people – theists, deists, atheists, and skeptics alike – can probably relate to that feeling of uncertainty and experience the onset of that all-important question: does God exist? Much of the debate about God’s existence focuses on the “teleological argument” – the argument from design. This argument suggests that God’s existence is the best explanation for the fine-tuning of the universe and the apparent design of things in the natural world. The teleological argument has unfortunately seized the focus from some of the more formidable philosophical arguments for theism – specifically the “cosmological argument.”

Cosmological arguments attempt to demonstrate the existence of a deity based on the mere existence of the universe. Many variants of the cosmological argument have been developed over the years and they fall into three rough categories: Kalaam, Thomistic, and Leibnizian.[i] Both the Kalaam and Thomistic arguments invoke a causal principle, which asserts that every event, or instance of something coming into existence, must have a cause.[ii] Kalaam and Thomistic arguments approach the problem of infinite causal series differently. The Kalaam argument maintains that because the past is finite, an infinite regress of causal events is impossible. Thomistic arguments do not deny the possibility of an infinite regress of causes, but as contemporary philosopher Alexander Pruss points out, they “use a variety of methods to argue against the hypothesis that there is an infinite regress of causes with no first cause.”[iii] Leibnizian arguments approach the question of existence differently. Rather than focusing on an infinite regress of causes, Leibnizian arguments rely on the Principle of Sufficient Reason (PSR), which holds that everything has an explanation, to argue that only a necessary being, i.e., one who does not rely on anything other than itself for an explanation of its existence, can provide an explanation for why the universe exists.

The Leibnizian Cosmological Argument from Contingency (LCA) makes a strong case for the existence of a necessary causally efficacious being. Before addressing the LCA, it is necessary to address a common objection to all cosmological arguments, which recurs in the literature of some of today’s more renowned atheists. In his work The God Delusion, Richard Dawkins discusses Thomas Aquinas’ “five ways,” several of which are variant forms of the cosmological argument, and proceeds to dismiss them as not proving anything because “they make the entirely unwarranted assumption that God himself is immune to the regress (of causes or extending back into the past).”[iv] Similarly, in God is not Great, Christopher Hitchens asserts that “the postulate of a designer or creator only raises the unanswerable question of who designed the designer or creator.”[v] Implicit in Dawkins’ and Hitchens’ remarks is the impression that all cosmological arguments require everything that exists to have a cause. If this were the case, then cosmological arguments would fail to show that God exists, because the very premise of an argument for the existence of God would require God to have a cause. It would then follow that God, since it required a cause, could not end a regress of causes extending back infinitely into the past.

Unfortunately for these critics, the premise – everything that exists has a cause – that Dawkins and many other New Atheists consider to be a fundamental feature of cosmological arguments is a straw-man since it is not defended by any serious proponent of the cosmological argument. The actual premise in essentially all cosmological arguments is rather that everything that begins to exist has a cause. Dawkins or Hitchens might retort, “What grounds do you have for claiming that God did not come into existence?” In short, a being who comes into existence would not be God. God, as understood by philosophers of religion, is a necessary being, one who contains the reason of its own existence within its very being, and therefore whose existence is self-explanatory, though probably not in a way that human beings can comprehend.

The LCA can be condensed from the abstruse shape it takes in Leibniz’s works into the following argument, a modified version of the cosmological argument presented by Jordan Howard Sobel in his Logic and Theism.

(1) The world – the cosmos – exists.

(2) The world is contingent; it is a contingent entity.

(3) For everything that exists, for every fact and every existent entity, there is a sufficient reason for its existence.

(4) There must be an explanation for the existence of the world. (1, 2, 3)

(5) The universe and everything within it, being contingent, cannot be explained by another contingent thing.

(6) The fact that there are contingent things must be explained by something whose existence is necessary. (4, 5)

Conclusion: Therefore, there exists an ultimate reason for the world, which is none other than a causally efficacious necessary being.v[i]

This argument is valid, which means if its premises are true, then its conclusion follows. (4) and (6) are dependent premises; that is, they do not stand on their own but rather follow from prior premises, and therefore do not need to be defended. (1) is intuitive and is therefore presumed to be true. The strength of the argument, then, hinges on its other three independent premises – premise (3), which dozens of philosophers object to, premise (2), which was critiqued by Bertrand Russell, and premise (5), which David Hume took issue with.

The third premise is Leibniz’s renowned Principle of Sufficient Reason (PSR), which affirms that for each thing that exists, there is sufficient reason – a complete explanation – as to why it exists. Needless to say, many philosophers have objected to the PSR for several reasons: 1) it purportedly lacks evidence, 2) it may entail the denial of contingency, 3) quantum mechanics appears to undermine it, and 4) if it holds, then it probably entails the existence of a causally efficacious necessary being.

Contrary to the first objection, “considered as an inductive generalization, PSR is as well supported as any other.”[vii] The world does not operate in a way that one would expect it to if the PSR were false; if this were the case, then “events without any evident explanation would surely be occurring constantly and the world would simply not have the intelligibility that makes science and everyday commonsense as successful as they are. This would be a miracle if the PSR were not true.”[viii] Furthermore, as Alexander Pruss notes, if the PSR were false, there might be no reason for our having the perceptual experiences we have.[ix] In other words, we would not have grounds to trust our perception if the PSR were not true because “objective probabilities depend on the objective tendencies of things, and if PSR is false, then events might occur in a way that has nothing to do with any objective tendencies in things.”[x]

The second objection may be true. Gottfried Leibniz did not hold that the PSR entailed the denial of contingency, but his fellow rationalist philosopher, Benedict de Spinoza, certainly did. Other philosophers, such as Peter van Inwagen and Jordan Howard Sobel, argued that the PSR implies modal fatalism – an extreme form of determinism that holds that everything happens necessarily. While the details of this argument and its rebuttals are beyond the scope of this article, Alexander Pruss successfully defends the PSR against Peter van Inwagen’s allegation that it implies modal fatalism in his book The Principle of Sufficient Reason.[xi] Given that the PSR does not necessarily imply modal fatalism, the third objection – that quantum mechanics disproves the PSR – can be dismissed. The very fact that quantum phenomena can be predicted with a high degree of probability lends itself to the fact that they do have an explanation, “since they presuppose and are made intelligible by the laws of quantum mechanics.”[xii] This brings us to the final objection to the PSR – that it implies the existence of a necessary being. This objection to the PSR is also lacking. Obviously, rejecting any principle or method that is well-attested by evidence because it leads to an unsavory conclusion is intellectually dishonest, and this objection can be dismissed on the grounds that it renders the project of conducting rational inquiry meaningless.

In his essay Why I am not a Christian, Bertrand Russell proposes that “there is no reason why the world could not have come into being without a cause; nor, on the other hand, is there any reason why it (the universe) should not have always existed.”[xiii] His first claim does not undermine the second premise of the LCA if the PSR is true. This is because the PSR denies the existence of brute facts – facts that lack an explanation – which means that if we grant the truth of the PSR, then the world could not come into existence without a cause. Russell’s second assertion may pose a threat to the LCA, since it suggests that the universe could exist necessarily, which would eliminate the need for an extramundane creator of the universe. How so? According to the Stanford Encyclopedia of Philosophy, “Russell contends that the move from the contingency of the components of the universe to the contingency of the universe commits the Fallacy of Composition, which mistakenly concludes that since the parts have a certain property, the whole likewise has that property.”[xiv]

Yet, reasoning from parts to the whole does not always involve committing the fallacy of composition. For example, consider a Lego house made entirely of green Lego bricks. One could infer, correctly, that the house is green given that each individual constituent brick is green. That being said, it is not certain that the inference from the contingency of the parts of the universe to the contingency of the universe itself is comparable to the inference in the Lego example above. Thomas Aquinas asserts in his Summa Theologica that “because every composite is posterior to its component parts, and is dependent on them… every composite has a cause, for things in themselves different cannot unite unless something causes them to unite.”[xv] Since the universe is a composite of the material things that it encompasses, or the events that take place within it, it must have a cause that explains how the components of the universe united to form the universe itself.

This results in two conclusions. First, because this cause logically precedes the formation of the universe, it cannot be identical to the universe. Second, because the “individual” components of the universe (e.g. mathematical laws, material objects, etc.) are not intrinsically prone to join together and form the universe (e.g. there is nothing inherent in math that requires it to exist alongside planets) the universe could not exist necessarily. Therefore, Russell’s objection fails. This raises another question, however, of whether a series of causes or reasons can sufficiently explain the existence of the series as a whole, a question that the fifth premise seeks to address.

The fifth premise does not claim that a contingent thing cannot explain another contingent thing. Rather, it claims that something contingent cannot explain why contingent facts exist at all. David Hume, in his Dialogues Concerning Natural Religion, challenged this premise by suggesting that once one has explained each contingent fact in a set of facts, one has sufficiently explained the set of facts as the whole. Therefore, since a contingent fact can explain another contingent fact, Hume deemed it plausible that there could exist a regress of events, contingent facts, each of which is explained by its predecessor, which would be fully explained by providing explanations for each member of the regress. Once you have explained the parts, you have explained the whole. As William Rowe points out in The Cosmological Argument, this line of reasoning, though probably correct in the case of some collections of contingent facts, does not work when applied to infinite regresses of causes or explanations.[xvi] Rowe offers the following example:

Consider M, the set of men. Suppose M consists of an infinite number of members, each member owing its existence to some other member which generated it. Suppose further that to explain the existence of a given man it is sufficient to note that he was begotten by some other man… Every member of M has an explanation of its existence, namely the causal activity of another man. But does it follow that the existence of M has an explanation? I think not. We do not have an explanation of the existence of M until we have an explanation of why M has the members it has rather than none at all. But clearly if all we know is that there always have been men and that every man’s existence is explained by the causal efficacy of some other man, we do not know why there always have been some men rather than none at all?….To explain the existence of the set we must explain why it has the members it has rather than none. But if every member’s existence is explained by some other member, then although the existence of every member has an explanation it is still unexplained why the set has the members it has, rather than none at all.[xvii]

Rowe’s argument essentially says that a sufficient explanation for an infinite regress goes beyond the individual explanations for each constituent member; there must be an explanation as to why the infinite set has the members it contains rather than other members, which cannot be explained by the causal activity of any member within the set. Even Jordan Howard Sobel, a staunch atheist, concurs with Rowe, noting that “Demea (one of the two theists in Hume’s dialogues) would have been right to insist that, even if shown the cause of each cause in infinite successions without going outside of them, one needs to go outside of them to find causes or reasons for these successions.”[xviii] Bearing this in mind, it is clear that Hume’s objection fails to present a serious challenge to the LCA.

Given the veracity of the PSR and the failure of Hume’s and Russell’s objections, the LCA succeeds in demonstrating the existence of a necessary being. Even if for some, the LCA is not convincing on its own, exploring these ideas can offer a new appreciation of one of the most prominent arguments for the existence of God. More importantly, charitably understanding the LCA cultivates a newfound caution when listening to skeptics who claim to understand (and dismiss) theistic arguments. The fact that cosmological arguments continue to be defended and debated attests not only to the nuance of the arguments presented, but also to the tremendous influence that cosmological arguments have had in elevating religious discourse.


i. Alexander R. Pruss, “Leibnizian Cosmological Arguments,” in The Blackwell Companion to Natural Theology (Oxford: Blackwell, 2009), 24.
ii. Pruss, 25.
iii. Pruss, 25.
iv. Richard Dawkins, The God Delusion (London: Bantam Press, 2006), 101.
v. Christopher Hitchens, God is not Great (New York: Hatchette Book Group, 2007), 71.
vi. Jordan Howard Sobel, Logic and Theism (New York: Cambridge University Press, 2004), 208.
vii. Edward Feser, Scholastic Metaphysics: A Contemporary Introduction (Neunkirchen-Seelscheid: Editiones Scholasticae, 2014), 142.
viii. Feser, 143.
ix. Pruss, “Leibnizian Cosmological Arguments,” 4.
x. Feser, 144.
xi. Alexander R. Pruss, The Principle of Sufficient Reason (New York: Cambridge University Press, 2006), 97-125.
xii. Pruss, The Principle of Sufficient Reason, 142.
xiii. Bertrand Russell, Why I Am Not a Christian (New York City: Simon & Schuster, 1967), 7.
xiv. Bruce Reichenbach, “Cosmological Argument,” The Stanford Encyclopedia of Philosophy, ed. Edward N. Zalta, 21 March 2013, <http://plato.stanford. edu/archives/spr2013/entries/cosmological-argument/>.
xv. Thomas Aquinas, Summa Theologica 1. 3. 7. <http://www.newadvent.org/summa/1003. htm#article7>.
xvi. William L. Rowe, The Cosmological Argument (New York: Fordham University Press, 1998), 154.
xvii. Rowe, 154-155.
xviii. Sobel, 216.


Luke Dickens ’18 is from Fortson, Georgia. He is a prospective Philosophy major.

Tags: , , , , , , , , , , , , , , , , , ,