Questions Does God exist?Whatever begins has a cause: the universe too

Does God exist?TheismCosmological

Whatever begins has a cause: the universe too

The kalam cosmological argument

William Lane Craig born 1949Revived the kalam argument: the universe began, so it has a cause.

What it claims

The universe has a cause (first step); and this cause, beginningless and outside time, is a personal agent who freely chose to create: a personal Creator (second step).

Why

Whatever begins has a cause; the universe began; its beginningless cause can only be an agent who freely chooses.

The same idea through history

Al-Ghazali1095In The Incoherence of the Philosophers and in al-Iqtisad, he argues against the philosophers of his day that the world began, and that whatever begins has a cause. He already adds that this cause must choose: it is a will.

Bonaventure1250In his commentary on the Sentences (book II, distinction 1), he holds that a world without a beginning is impossible: an infinite past could be neither traversed nor added to. Thomas Aquinas replies that reason cannot settle the question.

William Lane Craig1979He revives the argument under an Arabic name (kalam: the rational theology of Islam) and supports it with the mathematics of infinity and big bang cosmology.

In short

"Whatever begins has a cause: the universe too" is an argument by William Lane Craig (The Kalām Cosmological Argument, 1979; "The Existence of God and the Beginning of the Universe", 1991, 1979) on the question "Does God exist?" (answer: theism). Its claim: The universe has a cause (first step); and this cause, beginningless and outside time, is a personal agent who freely chose to create: a personal Creator (second step). According to AskPhilosophy's logic check (Z3 solver), the argument is valid only if one grants a premise the author leaves unstated (P5). AskPhilosophy presents 5 objections, 5 of them with a documented reply, and the original text behind each premise, checked word for word.

The Kalām Cosmological Argument, 1979; "The Existence of God and the Beginning of the Universe", 1991

Valid with an implicit premise

Things do not appear out of nothing: a house, a tree, a child have a cause. Now, says Craig, the universe itself began to exist. So it must have a cause. The argument comes in two steps. First step: the universe has a cause (P1, P2). For P2, Craig gives two reasons: an infinite past is impossible (there cannot really be an infinite number of things), and modern cosmology, in his view, points to a beginning. Second step, more disputed: this cause did not begin, it is outside time, and it must be a person who chooses, otherwise the universe would have existed forever. The idea comes from the Muslim theologians of the Middle Ages, above all al-Ghazali. Difference with Leibniz: here, there is no demand for a reason for everything, only a cause for what begins.

click a premise to reject it

If you accept every premise, logic forces you to accept the conclusion. So the debate is about the premises.

Checked by a logic solver: the premises cannot all be true while the conclusion is false.
Without the implicit premise: Without P5, the computer finds a world where the universe began and its cause did not, but where this cause is neither a mechanical sufficient condition nor a person: for example a non-personal cause that is only a necessary condition, or that acts by chance. The conclusion "personal Creator" no longer follows.

The first step (P1, P2, so the universe has a cause) is valid with nothing added: the whole debate is about the truth of P1 and P2. Craig even offers a more modest version of P1: "if the universe began, it has a cause". The second step is valid only with P5, which Craig does not defend separately: he assumes a cause is either mechanical or personal. Without P5, nothing forces the cause to be a person. This is where Morriston attacks.

Analogy. Hilbert's Hotel has infinitely many rooms, all occupied. A guest arrives: each guest moves from room n to room n+1, and room 1 is freed. The hotel was full, it takes one more guest, and it has no more guests than before. For Craig, absurdities like this show that a real infinite cannot exist, and so neither can an infinite number of past days. For the person: water frozen forever implies cold forever; but a man seated forever can decide to stand up. Only a will can produce an effect that begins from an eternal cause.

Objections

Each objection is a conversation: click to see who replies and where the debate stands.

J. L. Mackie 1982 · attacks premise 2

The absurdities of Hilbert's Hotel vanish once we understand Cantor's set theory: for an infinite collection, "the whole is greater than the part" simply no longer holds, and there is no contradiction in that. And an infinite past does not require "crossing" infinity from a starting point: from any past day, there is only a finite distance to today. Thomas Aquinas already answered Bonaventure this way.

Craig replies

Craig 1984

No one disputes that Cantor's theory is consistent in mathematics. The question is whether such an infinite can exist in the real world. Mackie does not remove the absurdities, he names the rule that produces them. And saying that each finite stretch of the past could be traversed does not prove that the whole infinite series could be: that is moving wrongly from the part to the whole.

Graham Oppy 1991

If Cantor's theory is consistent, then worlds with a real infinite are possible. Saying there is none in our world is no longer an a priori proof but a question of physics, which we cannot settle. As for rejecting a past with no first day, that is a prejudice against infinity, not an argument.

See the rest of the debate (1)

Craig 2008

Oppy has since devoted a whole book to infinity (2006). Yet he concludes there that every position on infinity has heavy costs and that mathematicians do not agree. Rejecting the real infinite therefore remains a reasonable position, and the kalam defender is not irrational.

The debate goes on (our summary) The debate remains open. Many philosophers judge the hotel paradoxes strange but not contradictory, and physics uses infinity everywhere (continuous space, for example). But even an opponent like Oppy admits that no position on infinity commands consensus. Hence the importance of Craig's second reason: cosmology.

Where the debate stands (our summary)Cantor makes a completed infinity coherent in mathematics; Craig answers that the question is its real existence, and Oppy that this belongs to physics. Craig notes that Oppy himself admits every position on infinity has heavy costs. The core disagreement: moving from the mathematical infinite to the real infinite. No consensus.

Sean Carroll 2014 · attacks premise 2

The physicist Sean Carroll, in his debate with Craig (New Orleans, 2014): that the universe began may be true, but we do not know. The big bang marks where our theories stop working, not necessarily an absolute beginning. Cosmologists build models with no beginning (for example a universe that "bounces", Aguirre and Gratton, 2003).

Craig replies

Craig 2014

The Borde, Guth and Vilenkin theorem (2003) shows that any universe which, on average, is expanding cannot be infinite in the past: it must have a beginning. This also holds for multiverse scenarios.

Sean Carroll 2014

The theorem assumes a "classical" spacetime, without quantum physics. It only says where our classical description stops working, not what really happened. Alan Guth himself, one of the three authors, thinks the universe is probably eternal.

See the rest of the debate (1)

Craig 2014

Suppose that before classical spacetime there was an era of quantum gravity. If that era had no beginning, it would have produced our classical universe either from eternity or never. So, if it existed, it is the beginning of the universe.

The debate goes on (our summary) Physicists remain divided: Vilenkin thinks the universe probably began, Guth that it is probably eternal. The theorem's authors themselves write that inflationary models need other physics to describe their past boundary. Science has not settled it, and Craig presents P2 as more probable than its denial, not as proven.

Where the debate stands (our summary)The Borde-Guth-Vilenkin theorem shows that a universe expanding on average has a finite past in the classical description; Carroll answers that it does not hold in quantum gravity, Craig that the quantum era would then be the beginning. Physicists remain divided: an open point.

J. L. Mackie 1982 · attacks premise 1

As Hume showed, we can perfectly well conceive of something beginning to exist without a cause. If it is nonetheless impossible, that needs proving, and the kalam defender does not prove it. Moreover, the believer accepts a God without a cause: why refuse the same to the universe?

Craig replies

Craig 1984

Being able to imagine something does not prove it is really possible. The principle is so obvious that it need not be proven by something less obvious: does anyone sincerely believe a tiger could appear uncaused, out of nothing, in the room where he is reading? And God did not begin: P1 only concerns what begins.

Graham Oppy 1991

Craig claims that God exists uncaused, unchanging and outside time. But then nothing in principle prevents saying the same of the universe. And an appeal to intuition settles nothing when the two sides do not share the same intuition.

The debate goes on (our summary) The defender answers that the universe did begin (P2), whereas God did not: the asymmetry holds if P2 is true. So everything still rests on P2.

Where the debate stands (our summary)The debate is about the status of P1: something self-evident that needs no proof (Craig) or an intuition that can be contested (Mackie, Oppy). The asymmetry with God holds if the universe began (P2). Core disagreement on the role of intuitions.

Sean Carroll 2014 · attacks premise 1

"Cause" is an everyday word, useful inside the universe, where there are laws and an arrow of time. It does not apply to the universe as a whole. A complete physical model, such as Hartle and Hawking's (1983, the "no-boundary" universe), is self-contained: it needs no outside cause, even if the universe has a first moment.

Craig replies

Craig 2014

If the universe could pop into being from nothing without a cause, why only universes? Why not bicycles, Beethoven or root beer? Nothingness does not choose. And a "complete" physical model describes the universe, it does not say why it exists.

Sean Carroll 2014

"Popping out of nothing" is a misleading image: it suggests there was a "before". It is better to say the universe has a first moment, with nothing before it.

See the rest of the debate (1)

Craig 2014

Changing the words does not change the thing. The job of philosophy is precisely to unpack such phrases and show what they imply: an uncaused first moment is still something coming from nothing.

The debate goes on (our summary) The disagreement is deep: for Carroll, physics settles what exists and needs no causes outside the universe; for Craig, metaphysics can ask questions physics does not. Morriston (2002) defends an idea close to Carroll's: our ordinary causes exist only within an already existing natural order.

Where the debate stands (our summary)For Carroll, physics settles what exists and a complete model needs no outside cause; for Craig, an uncaused first moment is still something coming from nothing. Core disagreement on what physics can say about the universe's existence; no consensus.

Wes Morriston 2000 · attacks premise 4

If God is outside time, his will to create is eternal, and that will is enough to produce the world. By P4, the world should then have existed forever. And if an eternal will can produce a world that begins, why not a non-personal cause? Outside time, the difference between a personal cause and a mechanical one disappears. Moreover, Craig elsewhere accepts causes that are only necessary conditions: so P5 (mechanical or personal, nothing else) is doubtful.

Craig replies

Craig 2002

Morriston confuses two things: God's eternal intention to create a world that begins, and the act by which he creates it. The intention is eternal, but the act of creating is free and coincides with the first moment. It is not one state mechanically producing another.

The debate goes on (our summary) Craig's reply assumes a theory of time in which the present really passes, and a freedom not reducible to causes. His opponents do not accept this. Even philosophers willing to grant that the universe has a cause often judge this second step weaker than the first.

Where the debate stands (our summary)Morriston judges that a timeless God cannot create a world that begins without his eternal will being enough to produce the world from eternity; Craig distinguishes eternal intention from the free act coinciding with the first moment. The reply assumes a theory of time and freedom that his opponents do not accept: an open point.

What today's voices say

Defend it

Wesley HuffHe argues that believing everything came from nothing is harder than believing the Bible's miracles: no big bang without a big banger.

Cliffe KnechtleHe argues that everything that begins has a cause, the universe began, so it has an uncaused cause.

William Lane CraigHis signature argument: the universe began to exist, so it has a cause. In 2009 he said an infinite past 'seems absurd'. In 2014 he backed the beginning with the Borde-Guth-Vilenkin theorem and thermodynamics; Carroll answered that quantum physics leaves the question open. Later he said the cause must be a personal, unembodied, free mind, that the argument does not prove it is morally good, and that an actually infinite past is metaphysically impossible though not strictly self-contradictory.

Frank TurekIn his 2008 debate with Hitchens he argued that the universe had a beginning, 'so it had a beginner'. He also argues that time cannot be infinite, or today would never have arrived.

Francis CollinsIn the Salon interview he cites the beginning of the universe (the Big Bang) as one of the things that point to a creator. He adds that the cause must be outside space and time.

Mohammed HijabHis reasoning in the Oxford Union debate asks whether something can come from nothing, in the manner of the kalam, but he prefers the contingency argument.