Did the Universe Begin? The Kalam Cosmological Argument

A question that refuses to stay small

Ask a child where the world came from and you will get a chain of "and before that?" questions that does not naturally end. The adult version is no less stubborn. Either the universe has always existed, stretching back through an unbounded ocean of prior moments, or it had a first moment — a beginning before which there was, quite literally, nothing physical at all. There is no third option, and almost everything we believe about reality hangs on which it is. If the cosmos is a self-sufficient, eternal brute fact, then it is the deepest thing there is, and the religious instinct that something stands behind it is a sentimental error. But if the universe began — if space, time, matter, and energy all switched on together a finite time ago — then the most natural question in the world becomes unavoidable: what brought it into being? The Kalam cosmological argument is a disciplined attempt to take that question seriously without cheating, and to follow it wherever it leads.

The argument is ancient, but it is not a museum piece. It sits today at the intersection of medieval theology, contemporary analytic philosophy, and front-line cosmology, and it has drawn the sustained fire of some of the sharpest atheist philosophers alive. That alone makes it worth mastering carefully — not as a debating trick, but as an honest inquiry into the most basic feature of the world we find ourselves in.

The argument, stated with care

In its modern form the Kalam is a model of economy:

Premise 1. Whatever begins to exist has a cause of its existence.
Premise 2. The universe began to existSee the evidenceThe evidenceSourceTwo independent lines. (1) The standard Big Bang model traces the expansion back to a finite past boundary some 13.8 billion years ago. (2) The 2003 Borde–Guth–Vilenkin theorem (“Inflationary Spacetimes Are Not Past-Complete”) shows that any on-average expanding spacetime is past-incomplete.Who holds itPowerful corroboration — not an airtight proof, and the essay says so. Vilenkin’s own conclusions are strong but measured, and one of the theorem’s co-authors, Alan Guth, is not persuaded the universe began (he is sympathetic to past-eternal models such as Carroll–Chen). The science corroborates a beginning of the universe as we know it; the philosophical arguments carry the premise independently..
Conclusion. Therefore, the universe has a cause of its existence.

The logic is valid: if both premises are true, the conclusion follows with the force of arithmetic. Notice what the argument does not say. It does not claim that "everything has a cause" — a careless formulation that would invite the obvious retort, "then what caused the cause?" It restricts itself to things that begin to exist. That precision is not a dodge bolted on after the fact; it has been built into the argument from its origins.

The word "kalam" is Arabic for "speech" or "discourse," and it came to name the tradition of medieval Islamic theology that pressed this reasoning hardest. The Persian theologian al-Ghazali (1058–1111) gave the argument its classic shape in The Incoherence of the Philosophers, formulating it in a sentence that still sounds modern: "Every being which begins has a cause for its beginning; now the world is a being which begins; therefore, it possesses a cause for its beginning."¹ Al-Ghazali was reacting against Muslim Aristotelians who, following the Greek master, held the universe to be eternal and beginningless.

But the deepest root runs earlier, to a Christian. In the sixth century, the Alexandrian philosopher John Philoponus turned Aristotle's own commitments against him. Aristotle had banned a completed, actually infinite quantity from the physical world; Philoponus, in works such as Against Aristotle, on the Eternity of the World (c. 530), argued that an eternal past would require exactly such a completed infinite of elapsed events — and so, by Aristotle's own lights, the world must have begun.² This finite-past reasoning passed into Islamic kalam, then into Jewish and Christian scholasticism, and finally fell dormant for centuries. Its modern revival is owed almost single-handedly to the philosopher William Lane Craig, whose 1979 monograph and decades of subsequent work — culminating in a 100-page chapter with the physicist James Sinclair in The Blackwell Companion to Natural Theology (2009) — rebuilt the argument with the tools of analytic philosophy and contemporary physics. Craig credits an earlier book with setting him onto the argument — Stuart C. Hackett's The Resurrection of Theism (1957) — and wrote his philosophy doctorate on the cosmological argument at Birmingham under the philosopher of religion John Hick.³

Premise 1: out of nothing, nothing comes

The first premise looks almost too obvious to defend, and that is its strength. The principle that being does not arise from non-being — ex nihilo nihil fit — is among the most secure intuitions we have. If things could simply pop into existence uncaused, out of nothing, it becomes inexplicable why we do not see it happening constantly and everywhere: why bicycles and Beethoven symphonies and herds of zebra do not flicker into being in midair. The fact that reality is orderly rather than a lottery of spontaneous somethings is itself evidence that coming-to-be requires a cause. It is worth naming someone who actually denied this of the universe itself, since a premise this secure can look like a straw target: in a 1993 exchange with Craig, the atheist philosopher Quentin Smith wrote that "the most reasonable belief is that we came from nothing, by nothing, and for nothing." Smith's own later work argues instead from quantum cosmology — which is where the serious scientific pressure on Premise 1 has come from since, and which this essay takes up below.⁴ Premise 1 is also confirmed throughout ordinary experience—though its deepest warrant is the metaphysical intuition just named, not the inductive track record alone.

The most serious scientific challenge points to quantum mechanics. In the quantum vacuum, "virtual particles" appear and vanish in ways that are not deterministically caused; doesn't this show that things can begin without a cause? The reply is precise, not evasive. The quantum vacuum is emphatically not "nothing." It is a real physical entity — a sea of fluctuating energy, governed by the laws of quantum field theory, with a rich structure that determines exactly which particles can emerge and how often. A particle arising from the vacuum is no more an instance of something-from-nothing than a wave arising from the ocean. The objection trades on an equivocation: "nothing" in the physicist's casual usage means empty space with its quantum fields, while "nothing" in Premise 1 means non-being — no space, no fields, no laws, no anything. Against that latter, genuine nothing, the principle stands unscathed.

Premise 2: the case for a beginning

The controversial premise is the second. It is defended along two independent lines — one philosophical, one scientific — which is part of what makes the argument robust: even if one line failed entirely, the other would remain.

The philosophical arguments against an infinite past

The first philosophical argument holds that an actual infinite — an infinite collection existing all at once in reality — cannot exist, because it generates absurdities. The mathematician David Hilbert illustrated the trouble with a famous thought experiment. Imagine a hotel with infinitely many rooms, all occupied. A new guest arrives; the manager simply moves the guest in room 1 to room 2, room 2 to room 3, and so on forever, freeing room 1 — so a "full" hotel takes another guest. Worse, infinitely many new guests can be accommodated the same way. And if the guests in all the odd-numbered rooms check out, infinitely many have left, yet exactly as many remain as before.⁵ These are not mere curiosities; they are signs, the argument runs, that an actually infinite number of real things cannot be instantiated in the concrete world. If so, the series of past events cannot be actually infinite — so the past is finite, and the universe began.

The second philosophical argument grants, for the sake of debate, that an actual infinite could exist, and still finds a problem: you cannot form an actual infinite by adding one member after another. The past is not a static set; it is built up successively, one event at a time. But no matter how long you go on adding moments, you never complete an infinite collection — you can always add one more. To arrive at the present moment by traversing an infinite series of prior moments would be to complete the uncompletable, like counting down from infinity and finally reaching zero. Since we have arrived at today, the series we traversed to get here must have been finite. Critics such as Morriston reply that this picture quietly smuggles in a starting point from which to count: an infinite past has no first moment to set out from, so the image of "counting down from infinity" may beg the very question at issue. The defender's answer is that the problem was never about a starting point: the worry is that an actual infinite cannot be completed by successive addition, one moment at a time, whether or not the series ever had a first member — and a beginningless past is precisely such an uncompletable series offered to us as already complete.

A striking contemporary sharpening of the case against an infinite past, developed by the philosophers Alexander Pruss and Robert Koons out of a paradox published by José Benardete in 1964, is the Grim Reaper paradox. Picture infinitely many “reapers,” each set to kill you at a fixed moment if you are still alive—one at 1:00, another at 12:30, another at 12:15, and so on without end, packed ever closer to noon. You cannot survive past noon, since some reaper would have struck; yet no reaper can be the one who strikes, because for any reaper there is always an earlier one who would already have acted. The scenario is flatly contradictory—which suggests that the actually infinite army of reapers it requires cannot exist in the first place. Pruss and Koons show that the same structure can be spread across an infinite past, giving a fresh reason to judge a beginningless series of events impossible.⁶

These arguments are contested, and a careful defender concedes that the actual-infinite premise is exactly where critics push hardest. The strongest reply is that Cantorian set theory shows the actual infinite to be perfectly coherent mathematically — Hilbert's Hotel is counterintuitive, not contradictory — so the defender's claim must be narrowed to this: an actual infinite cannot be concretely instantiated in the physical world. Whether that metaphysical claim holds, given the mathematical consistency of the infinite, is precisely the disputed hinge. The arguments are offered, then, not as knock-down demonstrations but as serious reasons, drawn from the nature of infinity and succession, to doubt that the past stretches back forever.

The scientific corroboration — stated honestly

Independently, modern cosmology points the same direction. The standard Big Bang model traces the expansion of the universe back toward a finite past boundary some 13.8 billion years ago. More striking is a result from 2003. The cosmologists Arvind Borde, Alan Guth, and Alexander Vilenkin proved a theorem — "Inflationary Spacetimes Are Not Past-Complete" — showing that any universe which has, on average, been expanding throughout its history cannot be extended infinitely into the past; its history must have a past boundary.⁷ Crucially, the theorem requires no assumptions about the energy content of the universe, which makes it far more general than earlier singularity theorems, and it applies even to most multiverse and eternal-inflation scenarios.

Here intellectual honesty is essential, because the argument is sometimes overstated. The Borde–Guth–Vilenkin theorem is a result about classical spacetime geometry. It establishes that the classical description breaks down at a past boundary; it does not, by itself, tell us what lies at that boundary — a genuine beginning, a quantum-gravity regime we do not yet understand, or something else. Vilenkin's own conclusions are strong but measured. In his book Many Worlds in One he writes that "with the proof now in place, cosmologists can no longer hide behind the possibility of a past-eternal universe. There is no escape, they have to face the problem of a cosmic beginning."⁸ In a later essay he is careful to say the answer to "Did the universe have a beginning?" is, "It probably did," adding that "we have no viable models of an eternal universe" — and, candidly, "a deep mystery remains."⁹ A 2012 paper with Audrey Mithani surveyed the three leading models proposed to avoid a beginning — eternal inflation, cyclic universes, and the "emergent" universe — and argued that none of them can in fact be past-eternal.¹⁰

And yet one of the theorem's own co-authors is not persuaded the universe began. Alan Guth has expressed sympathy for models — such as the Carroll–Chen proposal — on which the universe is past-eternal after all. The right conclusion, then, is sober: the science offers powerful corroboration that the universe as we know it had a beginning, not an airtight proof. This is precisely why the philosophical arguments matter independently — and why a careful case leans on convergence rather than on any single result.

A third consideration, the thermodynamic one, reinforces the picture. The second law of thermodynamics tells us that the universe's usable energy is steadily running down toward equilibrium — a uniform "heat death." But if the universe had already existed for an infinite time, it would presumably have reached that featureless equilibrium an infinite time ago. That it has not — that the cosmic clock is still winding down — suggests the clock was wound up a finite time ago. The details are genuinely disputed: in an expanding universe the maximum possible entropy rises over time, so some cosmologists, notably Sean Carroll, argue that equilibrium need never be reached. The thermodynamic consideration is therefore best treated as suggestive rather than decisive — but the low-entropy past we actually observe still sits awkwardly with a literally infinite history. Two further results reinforce the thermodynamic line. Cyclic or “oscillating” models—which try to spin an infinite past out of endless expansion-and-contraction cycles—face a classic objection resting on a result of Richard Tolman: entropy carries over from one cycle to the next, so the cycles grow larger going forward and must shrink toward nothing a finite time in the past, leaving a beginning intact—though modern cyclic models try to evade this by diluting the accumulated entropy as space expands.¹¹ And in 2013 the physicist Aron Wall proved a singularity theorem showing that, given the generalized second law of thermodynamics, a spatially infinite expanding universe must be past-incomplete—a result Wall argues may extend even into the quantum-gravity regime where the older, purely geometric theorems break down, since the semiclassical approximation it uses is invoked only where curvature is low. The condition is real and popular presentations drop it: if space is finite instead, the generalized second law requires only that there be a finite amount of entropy-producing process in the past, unless the arrow of time reverses—weaker than a past boundary, but not a return to a thermodynamically eternal history. That is the pattern across this whole section: each result can be escaped, and each escape carries a further commitment of its own—a rising entropy ceiling for Carroll, entropy dilution for the cyclic models, a reversed arrow of time here—each of which is itself debated. The thermodynamic line is convergent support, not a proof.¹²

From a cause to a Creator

Suppose the argument succeeds: the universe has a cause. What can be said about that cause, working purely from the premises, before importing any religious commitment?

A great deal, it turns out. The cause brought all of physical reality into being — all space, all time, all matter and energy. Therefore the cause cannot itself be spatial, temporal, or material; it must be spaceless, timeless, and immaterial, transcending the very framework it produced. It must be unimaginably powerful, having brought the entire cosmos into existence. And it must be uncaused — for if it began to exist it would, by Premise 1, require a cause of its own, and we have simply moved the question back a step. The cause of the universe is itself beginningless.

One further inference, more tentative but powerful, points toward a personal cause — a mind rather than a mechanism. Two considerations converge. First, whatever is timeless, spaceless, and immaterial seems to fall into one of only two categories we can conceive: an abstract object (like a number or a set) or an unembodied mind. But abstract objects, by their nature, stand in no causal relations — the number seven does not do anything. By elimination, the cause is plausibly a mind. Second, there is the puzzle of how a timeless cause produces a temporal effect at all. If the cause were a changeless set of sufficient conditions — like water freezing the instant the temperature drops below zero — then wherever the cause existed, the effect would exist too, and a timeless cause would yield an equally timeless, beginningless universe. But the universe is not beginningless. The most natural explanation for how a changeless, eternal cause gives rise to an effect with a beginning is a free act of will — an agent who can bring about a new state of affairs without any prior determining change. This is agent causation, and it is the signature of a person.

A further, independent line—pressed by the Oxford philosopher Richard Swinburne—converges on the same conclusion. Swinburne distinguishes two kinds of explanation: scientific explanation, which accounts for an event by earlier conditions together with the laws of nature, and personal explanation, which accounts for it by an agent acting for a purpose. Scientific explanation works by appeal to prior physical conditions and the laws governing them—which is why it can explain events within the physical order but cannot, without circularity, explain the existence of that order itself. A critic will answer that a whole needs no explanation once its parts are explained, or that the universe is simply a brute fact. That objection is the central battleground of the argument from contingency, where the same target is pressed by a different route and the objection is met at length; what Swinburne adds here is narrower. Note that he presses it independently of whether the universe had a first moment, so it is not a rider on Premise 2: his target is the existence of a complex physical universe as such. This falls short of proof, but it shows that the personal hypothesis is no arbitrary preference: on Swinburne's taxonomy of explanation, it is the only kind still standing once the scientific kind has run out.¹³

These last steps are best understood as inference to the best explanation, not deductive proof. One can resist them. But taken together they make a personal Creator the most reasonable account — a transcendent mind that freely brought the universe into being.

The strongest objections, fairly stated

An argument is only as good as its performance against its ablest critics. Here are the most serious, each in its strongest form.

"What caused God?"

The most common objection charges special pleading: if everything needs a cause, then God does too, and stopping at God is an arbitrary exception. But this misreads the first premise. Kalam never claims that everything has a cause; it claims that everything which begins to exist has a cause. God, by definition, did not begin to exist. So He is not an unexplained exception to a universal rule — He simply falls outside the scope of a rule that was about beginnings all along. The question "what caused God?" is, as defenders like to put it, rather like asking what the bachelor's wife is doing this evening: it presupposes what the case has already ruled out.

"Then isn't God infinitely old?"

A sharper cousin of that objection turns the argument's own philosophy back on it. If an actual infinite cannot be concretely instantiated, and if no infinite series can be completed by successive addition, what about God? And the objection has a serious form: some theists hold that God is everlasting in time rather than outside it, and on that view God's own past would be an unbounded series of moments, now complete — the very thing the two arguments were built to tell against. Does the argument's own philosophy rule out its own conclusion?

The two cases are not alike, and the two philosophical arguments the objection invokes show why. Both are pressed here against a series of events — a past which is not a static set but is built up successively, one event at a time. So what they give reason to doubt is an infinite elapsed past. And the first argument's claim, once narrowed as the defender must narrow it, is that an actual infinite cannot be concretely instantiated in the physical world — whereas the cause this argument reaches is spaceless, timeless, and immaterial, transcending the very framework it produced. On the classical account of divine eternity, a timeless cause is not an infinite series of moments completed, because it is not a series of moments at all＊Eternal, and not remoteIs sayingThat God's eternity is not a stretch of elapsed time that could be counted up or added to. On the classical account it is life possessed whole rather than lived moment after moment; Christians differ over how God stands to time, and the Kalam settles none of it.Not sayingNot that God is frozen, inert, or unable to act in time or be known. The God of the creeds acts throughout history, and in Jesus Christ the eternal Son took our nature and lived a human life — crucified "under Pontius Pilate." Eternity is what makes God present to every moment, not what removes him from them.. What the arguments tell against is an infinite elapsed past; what the argument concludes to is a cause that is no part of the physical world's past at all — and which, as the argument itself showed, did not begin to exist.

Two cautions belong with that reply. The defender should not answer by saying that God is "infinitely old," since that concedes the very picture being denied, treating God's eternity as a vast stretch of elapsed time. And how God stands to time is itself genuinely disputed among orthodox Christian philosophers and theologians — the classical tradition holds his eternity to be life possessed whole rather than lived moment after moment, while others hold that God's own life has some succession to it. The Kalam settles none of that, and this essay does not adjudicate it. On the classical account the objection dissolves, which is why the reply above is put that way. On the other account the defender owes a further answer about God's own past — but the Kalam's conclusion does not fall with it either way, since neither account makes God a member of the physical past these arguments are about, and Premise 2 never stood on the successive-addition argument alone. The same narrowing meets a cousin of the objection that presses on God’s knowledge rather than his past: if an omniscient mind grasps infinitely many truths at once, is that not an actual infinite? Perhaps it is — but the premise the defender actually needs is only that an actual infinite cannot be concretely instantiated in the physical world, and a mind’s knowledge is no part of the physical world. So the arguments no more forbid an infinite knowledge than they forbid an infinite Knower.

Wes Morriston on the causal principle

A far more sophisticated challenge comes from the philosopher Wes Morriston, a fair-minded and persistent critic. In "Must the Beginning of the Universe Have a Personal Cause?" Morriston questions whether we really have a rational intuition supporting Premise 1 as applied to the beginning of the universe itself.¹⁴ Our confidence that things have causes, he argues, is built up from experience of events within time and space — one thing emerging from prior materials in an existing world. The absolute beginning of all space and time is utterly unlike that. So why assume the principle reaches that far? Morriston also presses on the move to a personal cause, arguing that the very idea of a timeless agent "timelessly willing" a temporal first moment is harder to make coherent than defenders admit. This is the objection worth taking most seriously: the defender must argue that the causal principle is grounded not merely in inductive habit but in a metaphysical insight about being and non-being that holds regardless of whether the "beginning" is inside time or of time.

Graham Oppy and the brute-fact alternative

The Australian philosopher Graham Oppy, perhaps the most formidable contemporary atheist philosopher of religion, mounts a different attack. In Arguing about Gods and elsewhere, Oppy contends that the naturalist is under no rational compulsion to accept Kalam's conclusion.¹⁵ He questions whether the causal principle has the warrant its defenders claim, and he argues that even if the universe had a first state, the naturalist may simply take that initial state as a brute fact — something that exists without further explanation — or appeal to an impersonal natural cause rather than a mind. On Oppy's view, theism and naturalism each ultimately rest on some unexplained starting point, so theism enjoys no decisive advantage. The defender's reply is that a personal cause is a better stopping point than a brute, unexplained physical state precisely because it offers an explanation — an agent's intention — rather than a refusal to explain; but Oppy's challenge keeps the debate genuinely open and forces the theist to defend, not assume, the superiority of the personal hypothesis.

Quantum cosmology and the no-boundary proposal

A scientific objection holds that quantum cosmology dissolves the beginning altogether. The most famous version is the Hartle–Hawking "no-boundary proposal," in which, near the earliest era, ordinary time is replaced by a fourth dimension of space, so that the universe has no sharp boundary or first moment — it is finite in the past but without an edge, "like the surface of the earth" with no place that is the "beginning." Stephen Hawking drew the theological moral himself in A Brief History of Time: "if the universe is really completely self-contained, having no boundary or edge, it would have neither beginning nor end: it would simply be. What place, then, for a creator?"¹⁶ The reply has several strands. The model is speculative and depends on the contested device of "imaginary time," whose physical interpretation is disputed even among physicists. More importantly, even on the no-boundary model the universe still has a finite past and is not eternal; it still begins to exist, in the relevant sense, even if there is no first instant in the way we ordinarily picture one. And smoothing the geometry of the earliest moment does not by itself supply a cause for why there is a finite cosmos at all rather than nothing.

"The cause needn't be the God of theism"

A final objection is simply true, and a careful defender should say so plainly. Even granting everything, the argument delivers a spaceless, timeless, immaterial, enormously powerful, and plausibly personal cause of the universe. That is a long way toward theism — but it is not yet the triune God of the Nicene Creed, nor even a morally good or worship-worthy being. Kalam, on its own, is consistent with deism.＊A first cause is not the "deist's god"Is sayingKalam by itself reaches only a spaceless, timeless, immaterial, enormously powerful, and plausibly personal cause — one strand in a cumulative case that, with fine-tuning, moral value, and the resurrection, points to the personal, Triune God.Not sayingNot that the site teaches deism, or that God is merely a distant, impersonal "first cause" who set things going and withdrew. Scope-honesty about one argument isn't a claim about who God ultimately is. This is not a flaw; it is honesty about scope. No single argument of natural theology was ever meant to carry the whole weight. Kalam is one strand in a cumulative case, alongside arguments from fine-tuning, moral value, contingency, and ultimately the historical evidence for the resurrection. Its job is more modest and more foundational: to show that the universe is not a self-explaining brute fact, and that the door to a transcendent cause stands open.

The honest state of the debate

Where does the argument stand today? Among philosophers, it remains live and genuinely contested. Premise 1 is widely regarded as plausible, though critics like Morriston have raised real questions about its scope. Premise 2 is the battleground: the philosophical arguments against an actual infinite have able defenders and able critics, and the scientific evidence, while strongly suggestive of a beginning, is corroborative rather than conclusive — a point the more careful proponents readily grant. The conceptual step from cause to personal Creator is the most contested of all, and is best held as a reasoned inference rather than a proof. The Kalam has not ended the conversation; it has sharpened it, and forced naturalists to articulate exactly where they get off the train.

One more piece of honesty belongs to that picture, and it concerns the argument’s own history. It is easy to assume that the whole theistic tradition reasons as the Kalam does; it does not. Thomas Aquinas held that the world’s beginning in time cannot be established by reason at all — “by faith alone do we hold, and by no demonstration can it be proved, that the world did not always exist.”¹⁷ Maimonides, in the Jewish tradition, reasoned much the same way: the world’s creation is accepted on the authority of Scripture and prophecy, not established by demonstration. Aquinas still judged God’s existence demonstrable — only by arguments that do not turn on a temporal beginning. None of this refutes the Kalam, and its defenders answer that the philosophical case against an infinite past has been pressed much further since the thirteenth century. But it does mean the tradition divided at exactly this seam, and saying so is a matter of plain fairness: a case is not strengthened by conscripting authorities who expressly declined to make it.

What Kalam does and does not establish

Stated without exaggeration, the Kalam cosmological argument makes a serious, defensible case that the universe began to exist and therefore has a transcendent cause — one that is spaceless, timeless, immaterial, beginningless, enormously powerful, and most plausibly a personal mind. It does not, by itself, prove the existence of the God of any particular religion, and it does not silence every objection; the best critics remain in the field, and an honest advocate names them rather than waving them away. What it accomplishes is to put the deepest question back on the table and to show that "the universe just exists, inexplicably" is not the only rational option — and perhaps not the most reasonable one. For the believer, it is not the whole house but a load-bearing wall: a reminder, offered "with gentleness and respect" (1 Peter 3:15), that the heavens still declare something, and that the oldest question a child can ask may have the most consequential answer of all.