The Existence of God: an argument for God
Did the universe have a cause? The Kalam cosmological argument
The Kalam cosmological argument holds that whatever begins to exist has a cause, that the universe began to exist, and so that the universe has a cause. Its defenders, from medieval Islamic theologians to William Lane Craig, take that cause to be God. Its critics question both premises and the step from a cause to God.
- 13 minute read
- Last reviewed:
The argument, step by step
The argument, as its defenders state it:
- Premise 1Everything that begins to exist has a cause of its existence.
- Premise 2The universe began to exist.
- ConclusionTherefore, the universe has a cause of its existence.
- From a cause to GodNo scientific law can explain the universe's origin, since its laws are part of it, so the cause must be personal.
1Premise 1Everything that begins to exist has a cause of its existence.
For the argument
For the argumentNothing comes from nothing
William Lane Craig holds that the premise rests on the intuition that something cannot come from nothing. Experience agrees, he argues: we never meet things coming into being without a cause, and if the universe could, there would be no reason other things could not.1
Against the argument
Against the argumentThe premise is neither self-evident nor shown by experience
Graham Oppy does not share Craig's intuitions about the premise. Wes Morriston argues that it is not self-evident, and that finding causes for events within the universe gives no evidence that the universe as a whole needs one.1
Against the argumentSome quantum events may have no cause
Morriston and Quentin Smith argue that events at the quantum level may happen without a cause, so the premise is not known to hold everywhere.1
Reply, for the argumentQuantum events still have causal conditions
Craig holds that quantum events still have some necessary causal conditions. Alexander Pruss argues that a cause can bring about its effect without determining it.1
Reply, against the argumentNeeded conditions are not yet a cause
Morriston asks what makes a set of conditions a cause, if they are needed for an event but not enough to bring it about.1
Reply, for the argumentUnpredictable is not uncaused
Craig adds that not all physicists agree that quantum events are uncaused, and that their unpredictability may reflect the limits of what we can know rather than reality.1
2Premise 2The universe began to exist.
What both sides acceptCosmologists describe the universe as expanding: traced back in time, it grows ever denser, toward the state called the Big Bang. How the first instants unfolded, and whether anything came before them, is a matter of speculation and debate among physicists. In modern set theory, an infinite set can be matched member for member with one of its own parts, and mathematicians treat this as a defining feature of infinite sets, not a contradiction. The question in dispute is whether such a collection can exist among concrete things.1
For the argument
For the argumentAn actual infinite cannot exist
Craig argues that an actual infinite, a collection of definite members more numerous than any whole number, cannot exist in reality. A library with infinitely many books would be no smaller after books were taken away, which he holds is absurd. A past with no beginning would be such a collection, so the past had a beginning.1
For the argumentAn infinite cannot be built one step at a time
A second argument: a series formed by adding one event after another can never become actually infinite. If the past had no beginning, Craig argues, the present could never have arrived, since an endless chain of earlier events would first have to be completed.1
Reply, against the argumentA beginningless past needs no starting point
Morriston argues that in a past with no beginning, infinitely many events have always already passed, and each new event is simply added to them; the argument, he holds, assumes what it sets out to prove. Alex Malpass argues that counting down from infinity does not mirror counting up to it.1
Reply, for the argumentWhy the present arrived now
Craig answers that if the past were actually infinite, as many events would lie behind yesterday as behind today, so nothing would explain why the present arrived now rather than earlier.1
Reply, against the argumentAsking why now relies on a disputed principle
Morriston answers that the past just is the series of events that have already happened. Asking why the series arrived now, he argues, appeals to the principle of sufficient reason, which holds that every fact that could have been otherwise has an explanation, and he holds that principle is suspect and out of place in Craig's argument.1
For the argumentBig Bang cosmology
Craig and James Sinclair also point to the Big Bang. Traced back, they argue, the universe reaches a state of infinite density that marks the beginning of matter, energy, space, and time, with no earlier state before it.1
Against the argument
Against the argumentPuzzling is not incoherent
Bede Rundle grants that an actual infinite is puzzling but denies that it is incoherent: infinite sets simply behave differently from finite ones. Some mathematicians, the Stanford entry reports, argue that this holds for any infinite set, whether of numbers or of real events, and on that basis Smith argues that Craig assumes infinite collections must behave like finite ones.1
Reply, for the argumentPossible in mathematics is not possible in reality
Andrew Loke answers that what mathematics allows is not always possible in the real world: an equation can have minus two as an answer, but minus two people cannot carry a box home.1
Against the argumentAn endless future is just as puzzling
Morriston argues that an endless future raises the same puzzles as a beginningless past. If God decreed that two angels would praise him in turn forever, infinitely many praises would be said; if the puzzles do not rule that out, they cannot rule out an endless past.1
Reply, for the argumentThe past and the future differ
Craig answers that past events have actually happened and future ones have not, so an endless future is only potentially infinite. The difference holds, he argues, on a tensed view of time, which philosophers call the A-theory.1
Reply, against the argumentOn Craig's view, the past is no more real than the future
Yishai Cohen argues that Craig's principle assumes what is in dispute, and that on Craig's own view only the present exists, so past events are no more real than future ones.1
Reply, for the argumentWhat has not happened stays potential
Craig holds that what has not yet happened remains merely potential, even if God has decided or foreseen it.1
Against the argumentThe universe may have passed through many bangs
Since a phase before the Big Bang cannot be ruled out, some suggest that the universe may have passed through repeated cycles of expansion and collapse, perhaps infinitely many, so that Big Bangs have happened again and again.1
Reply, for the argumentOlder cyclic models face serious problems
Bruce Reichenbach, the author of the Stanford Encyclopedia's entry, replies that no known physical laws account for a series of collapses and re-explosions, and, citing Smith, that even such a universe seems to be finite, so the argument against an infinite series would still apply. Drawing on the physicists Paul Davies and Richard Tolman, he adds that entropy, a measure of disorder, would rise from cycle to cycle, making each cycle longer than the last.1
Reply, against the argumentThe newer cyclic model responds to the entropy problem
Responding to these problems, Paul Steinhardt and Neil Turok have proposed a cyclic model, drawing on string theory, in which the universe passes again and again from a big bang to a big crunch. In it, dark energy drives a phase of accelerating expansion that dilutes each cycle's entropy and returns the universe to a nearly empty state before the next cycle begins.1
Reply, for the argumentHow did the cycles begin?
Reichenbach notes that the amount of dark energy makes a return to an original state impossible, so the universe will end in a cold death rather than cycle. He adds that dark energy is inferred rather than confirmed, and that a more serious question is how a cyclic universe came to be in the first place.1
Against the argumentSome models have no beginning of time
In the "emergent" model of George Ellis and Roy Maartens, there is no beginning of time: the universe rests in a nearly static state for an infinite time before it starts to expand.1
Reply, for the argumentA static state would still need a cause
Reichenbach replies that such a model would not conflict with the Kalam argument, since a universe in a static state would still need a cause. Craig and Sinclair argue that the rival models of the universe's origin are not viable.1
3ConclusionTherefore, the universe has a cause of its existence.
What both sides acceptIf both premises are true, the conclusion follows. The dispute, as the Stanford Encyclopedia's entry sums it up, turns on whether the universe is infinite in time, whether a first existent would need a cause and of what kind, whether something like the universe can be finite in time and yet have no beginning, and, crucially, the nature of infinity and its connection with reality.1
For the argument
The sources used give no point here.
Against the argument
Against the argumentThe beginning was not an event
Stephen Hawking proposed that the universe is finite in the past but has no boundary in space-time, and so no beginning. Adolf Grunbaum argues that events come only from other events, and that since no time came before the Big Bang, it cannot have a cause.1
Reply, for the argumentA beginning without a first moment
Craig defines "begins to exist" so that no first moment is needed: a thing has a beginning if the time it has existed is finite. Other replies widen "event" and "cause" to allow a cause that does not act in time.1
4From a cause to GodNo scientific law can explain the universe's origin, since its laws are part of it, so the cause must be personal.
For the argument
For the argumentOnly a free choice explains a beginning
Craig argues that if the cause were a set of conditions that had always been in place, the universe would have always existed too. Since it began, its cause must be a personal agent who freely chose to create it.1
Reply, against the argumentWhy a person?
Morriston answers that if a timeless agent's intention to create were enough, the universe would be as old as the agent, so a personal cause gains no advantage.1
Reply, for the argumentA free choice, not only an intention
Against the argument
Against the argumentAgents need bodies
Rundle argues that a personal agent needs a body, and that actions take place in space and time.1
Reply, for the argumentAn intention can explain without a mechanism
Richard Swinburne holds that an event is fully explained by citing an agent, the agent's intention, and the agent's powers, even when we cannot say how the intention was carried out.1
Against the argumentA quantum vacuum could be the cause
Against the argumentThe cause need not be a person
John J. Park argues that the argument establishes at most an initial singularity outside the universe, nonphysical and uncaused, and that this is simpler than a personal God.1
Reply, for the argumentAbstract objects cannot make things happen
Loke answers that abstract objects do not by themselves make things happen in the concrete world, and cannot explain why one kind of beginning occurred rather than another.1
Where does the argument come from?
The argument takes its name from kalam, the tradition of reasoned theology in Islam.1 Medieval practitioners of kalam argued that the world had a beginning, and a study by Herbert Davidson traces part of their reasoning to John Philoponus (about 490 to about 570), a Christian philosopher of Alexandria.12 Al-Ghazali (about 1056 to 1111) argued in The Incoherence of the Philosophers that an eternal world is impossible, against Avicenna (Ibn Sina, about 980 to 1037) and al-Farabi, who held that it is eternal.2
Bonaventure (1221 to 1274) brought the argument into medieval Christian thought, and William Lane Craig gave it new life in modern philosophy, in a 1979 book titled The Kalam Cosmological Argument and many later writings.1 Believers in creation divided over it: Maimonides (1138 to 1204), the Jewish philosopher, and Thomas Aquinas (about 1225 to 1274), the Catholic theologian, both held that reason alone cannot settle whether the world began.341
Across traditions
The argument took its classic form in Islam and was argued over within Judaism and Christianity.13 Buddhist and Hindu thinkers argued about a first cause in their own terms, and the sources used for this page record no Buddhist or Hindu answer aimed at the Kalam argument itself.51 Latter-day Saint teaching holds its own view of creation, and a Latter-day Saint writer has answered evangelical critics who used the argument against it.678
In the early Buddhist texts, Richard Hayes writes in a study of Buddhist arguments against a creator, the Buddha treats the question of whether the universe had a beginning as a distraction from more pressing concerns.5 Later Buddhist philosophers in India argued against a creator, and several of their reasons bear on this argument. Vasubandhu (about 400 to 480) and most other Buddhist scholastics held that the world arises from the intentional actions of countless beings over beginningless time, not from a single cause.5
Vasubandhu argued that one unchanging God could not produce a world whose parts arise one after another, and that a creator who was himself the effect of earlier causes would mean admitting that the universe has no beginning, the Buddhist view, and then no creator is needed. Hayes adds that no one actually held that the creator was himself such an effect, and that Vasubandhu answered the idea out of a philosopher's thoroughness.5 Dharmakirti (about 600 to 660) argued that no unchanging being can create, since every action requires change.5
In Christianity, Aquinas took up the question in On the Eternity of the World (1271) and concluded that no demonstration is possible either way. God could have made a world that always existed, he held, and that the world began is known by faith, not by philosophical proof. Many of his contemporaries thought they could prove a beginning, and the dispute was one of the most bitter in the early universities.4
Followers of the Hindu devotional traditions, who worship Vishnu, Shiva, or the Goddess as supreme, believe that ultimate reality is a personal Lord (Ishvara), the creator of the universe, who also transcends it.9 Some Hindu teaching describes that creation as having no first beginning. Shankara, the Advaita Vedanta teacher who flourished in the eighth century CE, identifies God (Ishvara) as both the material and the intelligent cause of the universe, which God brings forth in a beginningless cycle of manifestation, sustenance, and dissolution.10 Hayes writes that most Indian theists pictured God setting eternal atoms in motion, and rejected creation from nothing as absurd.5
Among Hindu philosophers, the Nyaya school, which saw itself as a rational defender of Hindu theism, argued for God mainly from design, but a brief argument from dependence also appears in the work of the Nyaya thinker Udayana. Because the universe has parts that come into existence at particular times, it must have a cause, and God is the cause that depends on nothing.111 Critics objected that a God without a body cannot cause anything, and Nyaya thinkers replied that God could take on a body and need not create in the way humans do.1
In Islam, al-Kindi (about 800 to 870), the first self-identified philosopher in the Arabic tradition, had already argued from Philoponus that the world is not eternal. A finite cosmos can have only a finite time, he reasoned, and reaching the present would mean crossing an actually infinite number of past moments, which cannot be crossed.12 Muslim thinkers also disagreed about kalam's arguments. Ibn Rushd (Averroes, 1126 to 1198) rejected the arguments for God of the Ash'arite theologians, a school of kalam, judging them neither demonstrative nor persuasive to ordinary people, and argued instead from Aristotle's physics to a first cause.132
In Judaism, Saadia Gaon (882 to 942), a rabbi and philosopher, gave four proofs that the world was created: from finitude, from composition, from accidents, and from time. Harry Wolfson traced them to Islamic kalam traditions.14 Maimonides rejected the kalam proofs. He held that neither creation nor an eternal world can be demonstrated, but judged creation the better view, since it does less violence to scripture and since a God without free will could not give commandments.3
Latter-day Saint teaching differs from the view of creation that became dominant in Christianity. Joseph Smith taught that the material world has eternal roots and rejected creation out of nothing; in this teaching, God organized the universe from elements that already existed.6 The philosopher David L. Paulsen reads him as teaching that chaotic matter, intelligences (which Paulsen calls primal persons), and lawlike principles are as eternal as God, so that creation brings order out of chaos rather than something out of nothing.15
Paul Copan and Craig, writing as evangelicals, argued against this teaching in a 2002 essay and in their book Creation out of Nothing (2004), drawing on arguments that an actual infinite cannot exist and on Big Bang cosmology. Blake Ostler, a Latter-day Saint writer, replied in the FARMS Review in 2005 on the biblical and historical side of their case. He argues that the Bible teaches creation from a chaos that already existed, and that creation out of nothing first appeared in Christian thought late in the second century CE.7 In a separate reply published by FAIR, an organization that answers criticisms of Latter-day Saint doctrine, practice, and history, Ostler argues that neither of Copan and Craig's arguments about infinity applies to an infinite past, or is sound for one. Even if they were sound, he argues, they would not apply to theories in which our universe arises from a beginningless quantum vacuum or from earlier universes, which he presents as speculative possibilities rather than empirical science.8
Where does the debate stand?
Since the late twentieth century, the Stanford Encyclopedia's entry reports, theists and nontheists alike have generally been skeptical of cosmological arguments; among theists, Alvin Plantinga judged this kind of argument ineffective. Michael Martin, J. L. Mackie, Smith, Rundle, Morriston, and Oppy, among others, argue that no current version of the cosmological argument is sound or gives probable evidence, while Craig and Loke defend the Kalam's premises, drawing on several fields.1 The entry adds that philosophers on both sides now offer increasingly detailed arguments, and that any case for or against the second premise from cosmology meets an ever-changing field, since cosmological theories keep developing.1
Common questions
What is the Kalam cosmological argument?
It argues for God in two premises: whatever begins to exist has a cause, and the universe began to exist, so the universe has a cause. Its defenders add that no scientific explanation can account for the universe's very beginning, since its laws and starting conditions are part of it, so the cause must be personal. Critics question each step.
Does Big Bang cosmology show that the universe had a beginning?
Cosmologists describe the universe as expanding from a very dense early state, but how the first instants unfolded, and whether anything came before them, is still debated. William Lane Craig and James Sinclair take the Big Bang as support for a beginning; critics point to cyclic models, and to an emergent model with no beginning of time.
Why is it called the kalam argument?
Kalam is the tradition of reasoned theology in Islam. Medieval practitioners of kalam, among them al-Ghazali, argued that the world had a beginning because an infinite series of past events is impossible. Bonaventure brought the argument into Christian thought, and William Lane Craig gave it new life in modern philosophy, beginning with a 1979 book.
Read next
Sources
- Bruce Reichenbach. Cosmological Argument. Stanford Encyclopedia of Philosophy, 2026. Accessed 2026-11-20.
- Frank Griffel. al-Ghazali. Stanford Encyclopedia of Philosophy, 2026. Accessed 2026-11-20.
- Kenneth Seeskin. Maimonides. Stanford Encyclopedia of Philosophy, 2024. Accessed 2026-11-20.
- Robert Pasnau. Thomas Aquinas. Stanford Encyclopedia of Philosophy, 2022. Accessed 2026-11-20.
- Richard P. Hayes. Principled Atheism in the Buddhist Scholastic Tradition. Journal of Indian Philosophy 16 (1988), pages 5 to 28 (author's copy, University of New Mexico), 1988. Accessed 2026-11-20.
- Becoming Like God, Gospel Topics Essays. The Church of Jesus Christ of Latter-day Saints, 2016. Accessed 2026-11-03.
- Blake T. Ostler. Out of Nothing: A History of Creation ex Nihilo in Early Christian Thought. FARMS Review 17, number 2, pages 253 to 320 (full text as archived by Scripture Central), 2005. Accessed 2026-11-20.
- Blake T. Ostler. The Doctrine of Creation Ex Nihilo: A Response to Copan and Craig, Part 3: Do Kalam Infinity Arguments Apply to the Infinite Past?. FAIR (fairlatterdaysaints.org), Reviews of The New Mormon Challenge. Accessed 2026-11-20.
- John B. Carman. Bhakti. Encyclopedia of Religion (Gale), via Encyclopedia.com, 1987. Accessed 2026-10-29.
- Neil Dalal. Sankara. Stanford Encyclopedia of Philosophy, 2021, introduction and section 1. Accessed 2026-10-29.
- Matthew R. Dasti. Nyaya. Internet Encyclopedia of Philosophy, introduction and section 3. Accessed 2026-10-29.
- Peter Adamson. al-Kindi. Stanford Encyclopedia of Philosophy, 2024. Accessed 2026-11-20.
- Fouad Ben Ahmed and Robert Pasnau. Ibn Rushd [Averroes]. Stanford Encyclopedia of Philosophy, 2025. Accessed 2026-11-20.
- Sarah Pessin. Saadya [Saadiah]. Stanford Encyclopedia of Philosophy, 2003, sections 1, 2, 3, and 9. Accessed 2026-10-28.
- David L. Paulsen. Joseph Smith and the Problem of Evil. BYU Studies 39, number 1, page 53, section "The Logical Problem of Evil". Accessed 2026-11-20.