Why Is There Anything at All? Leibniz's Argument from Contingency

The Most Persistent Question

Sit with a child long enough and you will eventually hear the question that no amount of science seems to dissolve: but why is there anything at all? Not why this star or that galaxy, but why a universe — why being rather than sheer, featureless nothing. The German polymath Gottfried Wilhelm Leibniz (1646–1716) gave the question its classic form in 1714: "the first question we have a right to ask will be, Why is there something rather than nothing?"¹ He thought the question had an answer, and that the answer was God.

This essay examines the argument Leibniz built around that question — the argument from contingency, also called the Leibnizian cosmological argument. It is worth saying at the outset what this argument is not. It is not the kalam cosmological argument, which contends that the universe began to exist and therefore had a cause. The contingency argument is indifferent to whether the universe is finite or infinitely old. As William Lane Craig puts it, "the existence of an eternal multiverse can be granted with equanimity by Leibniz because it is irrelevant to his argument."² What matters is not when the universe came to be, but why it exists at all rather than not.

The Argument, Stated Carefully

Behind the argument lies Leibniz's Principle of Sufficient Reason (PSR): "no fact can be real or existent, no statement true, unless there be a sufficient reason why it is so and not otherwise."³ Things do not just happen to be the case for no reason whatsoever; reality is, at bottom, intelligible.

The argument's most widely discussed modern formulation comes from Craig, who distills it into three premises:⁴

1. Everything that exists has an explanation of its existence, either in the necessity of its own nature or in an external cause.
2. If the universe has an explanation of its existence, that explanation is God.
3. The universe exists.

From these it follows that the universe has an explanation (from 1 and 3), and that this explanation is God (from 2). A word on what "God" means in Premise 2: at this stage it denotes only a necessary, transcendent, non-physical ground of contingent reality — the identification of that ground with the personal God of theism is deferred to the cumulative case (and to the further argument below that such a ground must be a mind). The crucial distinction is between two ways a thing can exist. A contingent being is one that could have failed to exist — you, this planet, the hydrogen atom. Nothing about its nature guarantees its being; it depends on something else. A necessary being, by contrast, exists by its own nature — it could not have failed to be, and so requires no external cause. Premise 1 simply says that for any existing thing, one of these two accounts must hold. To deny it is to say some things exist for no reason at all — which, on reflection, is harder to believe than the premise.

Now apply this to the universe — meaning the whole of physical reality, all space, time, matter, and energy. Could the universe be a necessary being, existing by its own nature? That seems wrong. Its contents are paradigmatically contingent: a different array of particles, different laws, fewer galaxies, or none, all seem perfectly possible. As Craig notes, many philosophers reject the idea that the universe exists by metaphysical necessity.² But if the universe is not necessary, then by premise 1 it has an external cause — something beyond the entire physical order. And the cause of all space and time cannot itself be spatial or temporal; the cause of all contingent things cannot itself be one more contingent thing, on pain of needing a further explanation. The explanation must therefore be a transcendent, necessary being — the terminus Leibniz called God.

Why the Whole Needs Explaining

A natural worry arises here. Grant that each thing in the universe is explained by some earlier thing. Why think the collection needs explaining at all? Perhaps once every member is accounted for, the whole is accounted for too. This objection is old and serious, and it deserves a careful hearing rather than a quick dismissal — which is precisely where the critics come in.

The Strongest Objections

1. Hume and van Inwagen: against the PSR

The first and deepest line of attack targets the Principle of Sufficient Reason itself. In David Hume's Dialogues Concerning Natural Religion (published posthumously, 1779), the character Cleanthes presses the point: "Did I show you the particular causes of each individual in a collection of twenty particles of matter, I should think it very unreasonable should you afterwards ask me what was the cause of the whole twenty. This is sufficiently explained in explaining the cause of the parts."⁵ Explain each member of a collection, Hume's Cleanthes says, and the demand for an explanation of the whole is a confusion — like asking for the cause of the human race after you have accounted for each person.

A more formidable contemporary version comes from Peter van Inwagen. His argument is a reductio: suppose the PSR is true, and consider the conjunction of all contingent truths — call it the Big Contingent Fact. By the PSR it must have an explanation. That explanation cannot be a necessary truth (necessary truths entail only necessary consequences, but the Big Contingent Fact is contingent). Nor can it be a contingent truth, for then it would be one of the conjuncts of the very fact it is supposed to explain — explaining itself, which is absurd. So the PSR has no coherent way to explain the Big Contingent Fact. Worse, van Inwagen argues, a strong PSR would entail that every truth is necessary — a modal collapse in which nothing could have been otherwise.⁶ If the PSR destroys all contingency, the theist cannot use it.

Reply. The theist has two moves, and the better defenders use both. First, the argument need not rest on the strong PSR at all. Richard Gale and Alexander Pruss, in their 1999 paper "A New Cosmological Argument," reformulate the principle as a weak PSR, traceable to Duns Scotus: not "every fact has an explanation" but "every fact possibly has an explanation."⁷ This far more modest claim is enough to run a cosmological argument, yet it is immune to the modal-collapse worry, since merely possibly having an explanation does not necessitate anything. Second, the collapse objection assumes that to explain a contingent fact one must necessitate it. But explanation by free choice breaks this. If God freely chooses to create, the choice explains the world without making the world necessary — a free agent could have chosen otherwise. The explanans is necessary (God), but the explaining act is free, so contingency survives. Van Inwagen anticipates this and is unpersuaded; and the reply leans on a libertarian view of free will — itself a disputed metaphysics — so it arguably relocates the controversy rather than ending it. This is contested. But the defender's position is reasonable, not desperate.

2. Russell: "The universe is just there"

In the famous 1948 BBC radio debate, the Jesuit Frederick Copleston pressed the contingency argument on Bertrand Russell. Russell's reply has become a slogan for the naturalist: asked whether the universe is "gratuitous," he answered, "I should say that the universe is just there, and that's all."⁸ The cosmos, on this view, is a brute fact — it exists, full stop, and the demand for a further explanation is a category mistake. Russell pressed further with a charge of the fallacy of composition: "Every man who exists has a mother, and it seems to me your argument is that therefore the human race must have a mother, but obviously the human race hasn't a mother."⁹ Just because each part of the universe is contingent, it does not follow that the whole is.

One might also revive Hume's sharpest thrust: why not let the material universe itself be the necessary being? If we are allowed to posit one necessary stopping point, why must it be God rather than matter-energy?

Reply. The fallacy-of-composition charge does not always hold; some properties do transfer from parts to wholes (a wall built entirely of brick is itself brick). Whether contingency transfers is exactly the question — critics rightly insist it is what is in dispute — and the theist gives a reason it does: a whole composed entirely of contingent things, with no necessary member anywhere in it, has nothing within it that grounds its existence — and adding up things that each could have failed to exist does not produce something that could not have failed to exist. As for treating the universe as the necessary being: the trouble is that the universe does not look necessary. A necessary being must exist in every possible world and could not have been otherwise in any respect. But we can readily conceive the universe with different laws, different particles, or not at all. Its constituents are contingent through and through, and a being whose every part could have been different is a poor candidate for something that exists by the sheer necessity of its own nature.

3. The "Taxicab Fallacy"

Critics from Schopenhauer onward have leveled a tu quoque: the theist invokes the PSR to demand a cause of the universe, then suddenly dismisses it the moment we ask, "And what caused God?" The PSR, they charge, is "like a hired cab" — dismissed once it has carried you to your destination. If everything needs an explanation, so does God; if God can be the unexplained stopping point, why not the universe?

Reply. This misreads the argument. Premise 1 never said everything has an external cause; it said everything has an explanation either in an external cause or in the necessity of its own nature. God is not an arbitrary exception to the PSR — God satisfies the PSR, as the being whose explanation lies in his own necessary nature. The argument concludes to a necessary being precisely because a necessary being is self-explanatory. To ask "what caused the necessary being?" is to ask what caused something that by definition could not fail to exist — a confused request. The taxi is not dismissed; it arrives at its proper destination.

The most formidable critic: Graham Oppy

The sharpest contemporary opponent is the Australian philosopher Graham Oppy, and he grants more than Hume or Russell did. Oppy accepts that explanation must stop somewhere — but he argues the naturalist can take the stopping point to be the cosmos itself, or its initial state. He frames a dilemma: either every possible world shares an initial segment with the actual world, in which case the universe's initial state is necessary; or it does not, in which case that state is a brute contingent fact.¹⁰ Either way, naturalism has a terminus. And by a principle of parsimony, Oppy contends, whatever explanatory role the theist assigns to God, the naturalist can assign to nature itself — positing one fewer entity, and so winning on simplicity.¹¹

This is the objection that should be given the last word, because it is the strongest. The theist's reply is not a refutation but a weighing. If the cosmos's initial state is brute, then the PSR — even the weak PSR — has been abandoned at the crucial point, and the naturalist owes an account of why this state rather than any other. If instead the initial state is necessary, the theist asks whether a specific physical configuration — describable, quantifiable, seemingly capable of being otherwise — is really the kind of thing that could exist by absolute necessity, in the way a non-composite, non-physical ground might. Oppy can press back that "God" is no more obviously necessary than the cosmos, and that parsimony favors stopping sooner. The exchange does not end in a knockout. It ends in a judgment about which terminus is more plausibly necessary and more genuinely explanatory — and reasonable people, having understood the arguments, divide.

The Honest State of the Scholarship

It would be a disservice to claim more than the evidence allows. Among professional philosophers of religion the contingency argument is taken seriously — defended ably by Pruss, Gale, Robert Koons, and Joshua Rasmussen, and pressed hard by Oppy, J. L. Mackie, and others. The central battleground is the PSR. Few today accept the unrestricted strong PSR in van Inwagen's target form; the action has moved to restricted and weak versions, and whether those suffice is genuinely disputed. The free-will reply to modal collapse is promising but not decisive. The status of brute facts — whether the universe can simply be "just there" — remains, after two and a half centuries, a real fork in the road on which honest thinkers part ways.

What It Does, and Doesn't, Establish

Suppose the argument succeeds. What has been shown? That there exists a necessary, transcendent, self-explanatory ground of all contingent reality — the answer to Leibniz's question. That is a great deal. But it is not yet the full God of Christian worship. The argument, by itself, does not deliver a personal, good, providential Father; the inference to personhood (that the cause is a mind who freely chooses) requires further steps, and the inference to the God of Abraham, Isaac, and Jesus requires the rest of the case — moral, historical, and revelatory.

The contingency argument is therefore best understood as one strand in a cumulative case, not a lone proof. It does what philosophical arguments do at their best: it shows that belief in a necessary ground of being is not a flight from reason but a candidate answer to reason's most stubborn question. Leibniz's child still asks why there is anything at all. The honest reply is that the question is real, that "the universe is just there" is one possible answer, and that a necessary being is another — and that choosing between them is among the most serious things a thinking person can do. We commend the argument not as a hammer but, in the spirit of 1 Peter 3:15, as a reason for the hope within us, offered with gentleness and respect to those who weigh it and disagree.