Berean Road

Choose a book

Topics

The Kalam Cosmological Argument

In plain English

The Kalam argument asks a simple question: did the universe have a beginning, and if so, what follows? Things that begin to exist don’t do so for no reason, and the universe appears to have begun. So it has a cause beyond space, time, and matter. If the argument works, it points to something much like the Creator of Genesis 1:1. Thoughtful people disagree about whether it works, and not only skeptics: Thomas Aquinas believed the world began but denied that reason could prove it. Below we lay out both sides, with sources you can check.

Key terms

  • Potential vs. actual infinite: A potential infinite can always go one more step but is never finished, like counting. An actual infinite is a completed collection with infinitely many members.
  • A-theory vs. B-theory of time: On the A-theory, time really passes and the future isn’t yet real. On the B-theory, past, present, and future are equally real, like points on a line.
  • Classical spacetime: Space and time as Einstein’s general relativity describes them, without quantum effects. It is expected to fail at the extreme densities of the earliest universe.
  • Past-eternal / past-incomplete: A past-eternal universe has no first moment. A spacetime is past-incomplete if the paths of particles or light, traced backward, run out after a finite time. Our description stops there, which doesn’t automatically mean the universe began there.

The argument

The modern form comes from philosopher William Lane Craig, building on medieval thinkers:

Premise 1: Whatever begins to exist has a cause.

Premise 2: The universe began to exist.

Conclusion: Therefore, the universe has a cause.

The logic is valid, so the debate is about whether the premises are true, and what kind of cause the conclusion points to.

Check it: Stanford Encyclopedia of Philosophy, “Cosmological Argument,” §8 · Craig, “The Kalam Cosmological Argument” (2015 Birmingham lecture)

Premise 1: Whatever begins to exist has a cause

Why believe it?

Craig gives three reasons. First, something cannot come from nothing, and “nothing” isn’t empty space but the absence of anything at all. Second, if things could pop into being uncaused, why only universes and not, in Craig’s example, bicycles or root beer? Third, experience and science both assume that beginnings have causes.

Note that the premise does not say “everything has a cause.” It covers only things that begin to exist.

Check it: Craig, Reasonable Faith lecture, “Premise 1”

Objection: Quantum physics shows things happen without causes

Events like radioactive decay seem to happen without a deterministic cause. Some philosophers, such as Quentin Smith, have argued that quantum mechanics limits the causal principle so much that we can’t confidently apply it to the Big Bang.

Response: Quantum events don’t happen in nothing, but in a setting of fields, energy, laws, and spacetime. As physicist Christopher Isham puts it (quoted by Craig), particle “creation” in physics is “the conversion of one type of matter into another.” Craig and James Sinclair also argue that “most of the available interpretations” of quantum mechanics are fully deterministic.

Still debated: The Stanford Encyclopedia is candid: is quantum indeterminacy real, or a limit on what we can know? If real, the causal principle is weakened.

Check it: SEP §8.1, “The Causal Principle and Quantum Physics”

Objection: Causation only applies within the universe

Cosmologist Sean Carroll, who debated Craig in 2014, argues that Premise 1 is “flat-out false. Or not even false,” because “the notion of a ‘cause’ isn’t part of an appropriate vocabulary to use for discussing fundamental physics.” Physics describes unbreakable patterns, not external causes; causes make sense inside a universe, not for the universe as a whole. A self-contained model, such as the Hartle–Hawking “no-boundary” proposal, is all the explanation needed.

Response: Defenders reply that “nothing comes from nothing” is a metaphysical principle science presupposes, not a lab discovery. Robert Koons, against Hume and Russell, argues we don’t need to watch a universe being caused to know it needs a cause, any more than we must watch every dropped ball to know it will fall.

Still debated: Wes Morriston replies that while we needn’t observe every case, the universe is “a very different thing” from the ordinary objects our causal intuitions come from. This may be the deepest disagreement of all: is “cause” a basic feature of reality or a useful shorthand? Carroll himself called his argument on this point “the most important part of the debate.”

Check it: Sean Carroll, “Post-Debate Reflections” (2014) · SEP §5.2 and §5.4 on Russell, Hume, Koons, and Morriston

Premise 2: The universe began to exist

Defenders support this premise in two independent ways: philosophically and scientifically.

The philosophical case: can the past be infinite?

Argument 1: An actual infinite can’t exist in reality. Craig argues that an actual infinite can exist in mathematics but not in the real world, because it produces absurdities.

One of his favorite illustrations is Hilbert’s Hotel, after mathematician David Hilbert. Picture a hotel with infinitely many rooms, all full. A new guest arrives, so everyone moves up one room and room 1 is free. Infinitely many new guests arrive, so everyone doubles their room number and the odd rooms open up. Full, yet always room for more. Craig argues this shows a concrete actual infinite is impossible, not just strange.

Sub-premise A1: An actual infinite cannot exist.

Sub-premise A2: A beginningless series of past events would be an actual infinite.

Sub-premise A3: Therefore, a beginningless series of past events cannot exist.

Argument 2: You can’t build an infinite one step at a time. The past was formed by successive addition, one event after another. You can never count up to infinity, so how could you finish counting down from infinity to reach today? Craig imagines a man finishing a countdown from eternity: why today and not yesterday, when he’d already had infinite time?

Check it: Craig, Reasonable Faith lecture, “First” and “Second Philosophical Argument” · SEP §8.2–8.3

Objection: Modern mathematics handles infinity just fine

Since Georg Cantor, set theory has handled actual infinites rigorously. Critics like Quentin Smith and Bede Rundle say the “absurdities” are just how infinite sets behave; calling them absurd brings finite intuitions into an infinite case.

Response: Craig agrees the math is consistent, but argues that consistency in a mathematical system doesn’t show something can exist in the physical world. Andrew Loke’s analogy: x² = 4 has two solutions, but if the question is “how many people carried the box home,” −2 isn’t an option.

Still debated: Wes Morriston argues the same logic would rule out an endless future, such as angels praising God forever. Craig answers that the future is only a potential infinite, a reply that depends on the A-theory of time.

Objection: An infinite past doesn’t need a starting point

Morriston and Alex Malpass say the countdown puzzle begs the question. In a beginningless past, every past event is only finitely far from now; as Morriston puts it, infinity is “always already there.” This isn’t only a skeptic’s point. Aquinas answered the same “infinite days” argument centuries ago: “Whatever bygone day we choose, from it to the present day there is a finite number of days which can be passed through.” And, as Adolf Grünbaum argued, on the B-theory of time the past isn’t “formed” by adding events at all.

Response: Loke argues his version of the argument works on either theory of time. Malpass grants there is “something rather strange” about always having been counting but never having begun.

Still debated: Malpass adds that turning that strangeness into a principle (“one cannot complete something unless one started to do it”) “would be begging the question” at this point in the debate.

Check it: SEP §8.2–8.3 on Rundle, Smith, Loke, Morriston, Malpass, Grünbaum · Aquinas, Summa Theologiae I q.46 a.2, reply to objection 6

The scientific case

The expanding universe. Run cosmic expansion backward and the universe gets ever denser. ESA’s Planck mission puts its age at 13.8 billion years. In the classical (general-relativity) Big Bang model, space and time themselves have a boundary in the finite past. Whether the real universe has such a boundary depends on physics we don’t yet have (see the caveats below).

Check it: ESA, “Celebrating the legacy of ESA’s Planck mission”

The Borde–Guth–Vilenkin (BGV) theorem. In a 2003 paper, physicists Arvind Borde, Alan Guth, and Alexander Vilenkin argued that “a cosmological model which is inflating – or just expanding sufficiently fast – must be incomplete in null and timelike past directions.” The result “depends on just one assumption”: that the expansion rate, averaged along a path traced back into the past, is greater than zero.

Strictly, that is past-incompleteness: some paths, traced backward, run out. It is not proof of a singularity or an absolute beginning. The paper’s own conclusion is that “inflationary models require physics other than inflation to describe the past boundary of the inflating region of spacetime.”

Vilenkin himself reads it strongly. In Many Worlds in One (2006, p. 176) he wrote that “cosmologists can no longer hide behind the possibility of a past-eternal universe.” But he dismisses the view that a beginning is evidence for God as “far too simplistic” (p. 177), a judgment Craig disputes.

Later work has questioned and refined the theorem. Lesnefsky, Easson, and Davies (2023) argue that the incompleteness claim “is an open issue,” partly because the average expansion rate can be calculated in different ways. They construct expanding models that are geodesically complete, though these require briefly violating standard energy conditions. Geshnizjani, Ling, and Quintin (2023) stress that incompleteness alone “does not tell us anything about the nature of the spacetime singularities,” and that some such spacetimes may be extendible. Still, they conclude that “past-eternal inflationary scenarios are most likely physically singular, except in situations with very special initial conditions.”

Check it: Borde, Guth & Vilenkin, arXiv:gr-qc/0110012 · Vilenkin quotes, via Craig’s review in Philosophia Christi · Lesnefsky, Easson & Davies, Phys. Rev. D 107, 044024 (2023), arXiv:2207.00955 · Geshnizjani, Ling & Quintin, JHEP 10 (2023) 182, arXiv:2305.01676

Thermodynamics. Systems left alone run down toward equilibrium. An infinitely old universe should already have reached that “heat death”; ours hasn’t.

Honest caveats: what the science does and doesn’t claim

  • BGV is about classical spacetime. It shows our classical description has a past boundary, not what lies at or beyond it. Carroll calls it “only a guideline to when our classical description breaks down.” The paper itself mentions options such as a “quantum nucleation event” and a time-reversed model by Aguirre and Gratton.
  • The hot, dense early state may not be the absolute beginning. Carroll, answering a reader in the comments under his post: the observable universe emerged from that state, “but we don’t know whether that state was really the beginning.”
  • Cosmology keeps changing. The SEP warns of “an ever-changing scene” and “unending point-counterpoint.” Thermodynamics is contested too: Carroll argues the Second Law reflects conditions in our region, not a bar to eternal models.

Fairly stated: the philosophical arguments, if sound, settle the question whatever physics says. Defenders see the science as strong but provisional confirmation. Critics disagree: Carroll calls Premise 2 “completely uncertain,” and Carroll reports that Guth, a BGV author, thinks the universe “probably didn’t have a beginning, and is very likely eternal.”

Check it: Carroll, “Post-Debate Reflections” · SEP §8.4, “The Big Bang Theory of Cosmic Origins”

Objection: Quantum mechanics implies an eternal universe

In the debate, Carroll offered what he calls the “Quantum Eternity Theorem.” As he later summarized it, under conventional quantum mechanics, “any universe with a non-zero energy and a time-independent Hamiltonian will necessarily last forever toward both the past and the future.” Because it takes quantum mechanics seriously, he argues, it’s a better guide than BGV.

Response: Craig replies that the theorem “allows us to describe a moment prior to a given moment if there is such a moment.” It doesn’t tell us whether there is one, and for that we need empirical evidence. Its assumptions may also fail. Carroll grants that “you could always violate the assumptions,” and notes that the Hartle–Hawking model “is not eternal to the past” because it assumes zero total energy. Craig reports that the physicists he consulted say most quantum cosmologists think the relevant energy is “zero or undefined.”

Still debated: Whether the universe’s total energy is zero, and what time even means in quantum gravity, remain open.

Check it: Carroll, debate transcript, “God and Cosmology” · Carroll, “Post-Debate Reflections” · Craig, “Does Quantum Mechanics Indicate an Eternal Universe?” (Q&A #389)

The conclusion: what kind of cause?

If the universe has a cause, Craig argues it must be:

  • Uncaused and beginningless, since an infinite chain of causes is ruled out;
  • Beyond space and time, since it brought space and time into being;
  • Immaterial, since it brought matter into being;
  • Enormously powerful;
  • Personal. This is the most debated step. A timeless, impersonal set of conditions would produce its effect from eternity, as water below freezing would always have been frozen. An effect with a beginning from a timeless cause is best explained by a free agent who chooses to create.

What it doesn’t prove. Alone, the Kalam doesn’t show that this cause is all-good or became incarnate in Jesus of Nazareth; those claims rest on other evidence, such as the case for the resurrection. But the profile it sketches (a transcendent, eternal, immaterial, powerful, personal Creator) fits what Scripture claims (Genesis 1:1, Psalm 90:2, John 1:1-3).

Check it: Craig, Reasonable Faith lecture, “Conclusion” · SEP §8.6, “Personal Explanation”

Strongest objections and responses

Objection: “Then who caused God?”

The serious version is a charge of special pleading: why may the theist stop the chain of explanation at an uncaused God, while the naturalist may not stop at the universe? As the Stanford Encyclopedia frames the challenge: if theists accept God as a brute fact, “why cannot nontheists accept the existence of the universe as a brute fact, as a necessary being?” J. L. Mackie adds that if God’s existence isn’t logically necessary, we can still ask, “Why, then, does God exist?”

Response: Premise 1 applies only to things that begin, so the chain may stop at whatever has no beginning. If Premise 2 is right, the universe doesn’t qualify; a beginningless cause does. Theists such as Stephen Davis and John Hick add that God is not the kind of being whose existence needs a further reason: “a demand for explaining its existence is inappropriate.”

Still debated: Carroll answers that even if “the universe has an earliest moment of time,” “a self-contained physical model is all you need.” So the question returns as a dispute over Premise 1.

Check it: SEP §5.2 and §5.5 · Carroll, “Post-Debate Reflections”

Objection: A timeless cause can’t be “personal” in any meaningful way

Morriston argues that if a timeless agent’s eternal intention to create was enough, the universe should also exist from eternity, erasing the personal/impersonal distinction.

Response: Craig replies that an agent must not only intend but exercise causal power, and that God can be timeless without the universe and enter time at creation.

Still debated: Morriston answers that exercising power is itself an act in time.

Check it: SEP §8.6

Objection: The Big Bang isn’t an “event,” so it can’t have a cause

Adolf Grünbaum argued there was no time “before” t = 0, so there is no first event to be caused.

Response: Craig argues that beginning to exist doesn’t require an earlier time, or even a first instant. He defines “x begins to exist” as “x exists at t and there is no time immediately prior to t at which x exists.” With Sinclair he adds: “Something has a beginning just in case the time during which it has existed is finite.” The SEP adds a puzzle: if the singularity isn’t an event and events only come from events, nothing after it could be an event either.

Check it: SEP §8.5, “The Big Bang Is Not an Event” (quoting Craig 1992 and Craig & Sinclair 2009, p. 185)

A short history

  • Al-Ghazali (1058–1111), the Persian Muslim theologian, argued in The Incoherence of the Philosophers against Greek-influenced thinkers who held the world was eternal. Craig quotes his summary from a different work, al-Iqtiṣād fī al-Iʿtiqād: “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.” The approach drew on the mutakallimūn and the Christian John Philoponus.
  • Bonaventure (1221–74), the Franciscan theologian, brought the argument into medieval Christian thought in his commentary on the Sentences.
  • Thomas Aquinas disagreed. “By faith alone do we hold, and by no demonstration can it be proved, that the world did not always exist,” he wrote. He warned that weak proofs give “occasion to unbelievers to laugh,” which is still good advice for apologists.
  • William Lane Craig revived and named it in his doctoral work under John Hick at the University of Birmingham, leading to The Kalām Cosmological Argument (1979). (“Kalam” is Arabic for medieval Islamic theology.) Leading critics include Graham Oppy, Wes Morriston, and Quentin Smith; Andrew Loke is a recent defender.

Check it: SEP §1, “Historical Overview” · Craig, Reasonable Faith lecture · Aquinas, Summa Theologiae I q.46 a.2

Scripture connections

The Bible doesn’t state the Kalam as a syllogism, but it teaches what the argument concludes: the universe had a beginning, and God, who has none, is its cause.

  • Genesis 1:1: the universe has a beginning, and God is its author.
  • Psalm 90:2: before the mountains and the world, from everlasting to everlasting, God is God.
  • John 1:1-3: the Word was already there at the beginning, and all things were made through Him.
  • Hebrews 11:3: by faith we understand that the universe was formed at God’s command, and that what is seen wasn’t made from what is visible.
  • Romans 1:20: God’s eternal power and divine nature are understood from His workmanship.

Check it yourself

Overviews (balanced)

The case for

The science

Critical voices

Berean Road’s approach: read the strongest version of each side, then check the sources, just as the Bereans did (Acts 17:11).