Questions Does God exist?The necessary existent: the proof of the truthful

Does God exist?TheismCosmological

The necessary existent: the proof of the truthful

The proof of the truthful (burhan al-siddiqin)

Avicenna c. 980 · c. 1037The soul knows it exists before any body: the flying man.

What it claims

There exists an existent necessary in itself, one, the cause of everything else: God.

Why

Whatever is merely possible gets its existence from a cause; the whole of such beings needs a cause outside it, necessary in itself.

The same idea through history

Avicenna1027In the Shifa' and the Najat, then in Pointers and Reminders (c. 1030-1034), where he deals in detail with an infinite chain of causes.

Thomas Aquinas1270The third way starts from things that come to be and pass away, and goes through the idea that at some time nothing would have existed: a detour Avicenna did not need to make.Gottfried Wilhelm Leibniz1714Leibniz starts from the principle of sufficient reason: even an infinite series of states of the world requires a reason outside the series.

In short

"The necessary existent: the proof of the truthful" is an argument by Avicenna (The Book of Salvation (Najat), "Proof of the necessary existent"; Pointers and Reminders, namat 4; Metaphysics of the Shifa', I, 6 and VIII, 3-4, 1027) on the question "Does God exist?" (answer: theism). Its claim: There exists an existent necessary in itself, one, the cause of everything else: God. According to AskPhilosophy's logic check (Z3 solver), the conclusion does follow from its 6 premises. AskPhilosophy presents 5 objections, 4 of them with a documented reply, and the original text behind each premise, checked word for word.

The Book of Salvation (Najat), "Proof of the necessary existent"; Pointers and Reminders, namat 4; Metaphysics of the Shifa', I, 6 and VIII, 3-4

Valid

Avicenna starts neither from motion, nor from the order of the world, nor from its beginning: he starts from the mere fact that something exists. Everything that exists, taken in itself, either exists necessarily in itself, or could fail to exist. What could fail to exist owes its existence to something else: I can know what a triangle is without knowing whether one exists, so its essence is not enough to make it exist. Its great advantage over the kalam argument (al-Ghazali, Craig): it does not need the world to have begun. Even an eternal world, or an infinite chain of causes, would need a necessary existent. Aquinas (third way), who had read Avicenna, and later Leibniz (sufficient reason) would follow a similar path, each in his own way. Avicenna calls it the proof "of the truthful", because it goes from existence itself to God, without passing through creatures.

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.

The structure is valid and needs no hidden premise. Either something is already necessary in itself, and we are done; or everything is possible, and then the whole of all possibles requires a cause that can only be necessary. It is a reductio ad absurdum. The attacks target three points. P4: does a whole each of whose members has a cause need a further cause? (al-Ghazali, Hume). P3: is existence really something "added" to essence? (Averroes). P6: from "a necessary being exists" to "it is God, one and perfect" is a long way, which Avicenna covers in other chapters (Kant, Hume).

Analogy. A room is lit through a set of mirrors. Each mirror reflects light it receives from another. Even with infinitely many mirrors, somewhere there must be a source that shines by itself, or no mirror would shine. The mirrors are the possible things; the source is the necessary existent.

Objections

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

Al-Ghazali 1095 · attacks premise 4

"Possible" just means "having a cause". In an infinite chain, each member has a cause: the previous member. But the whole has no further cause, external to it. What is true of each member is not necessarily true of the whole: each revolution of the heavens has a beginning, but, for the philosophers themselves, the whole series has none.

Avicenne replies

Avicenne

A whole each of whose members is caused requires a cause outside its members. If it required no cause, it would be necessary in itself; but how could it be, when it exists only through its members? And if its cause were all its members taken together, it would be its own cause, for that is the same thing.

The debate goes on (our summary) This answer was written before the objection (Avicenna died in 1037, al-Ghazali wrote in 1095), but it targets exactly this point. Al-Ghazali knows this argument: he reports it, and answers that saying "a necessary whole made of possibles is absurd" assumes what needs proving (The Incoherence, discussion 4). Averroes, though a philosopher, partly agrees with him: in his view Avicenna's fault lies in the way he divides existents, even though the proof can be corrected. Hume (1779) and then Russell (1948) would revive the objection: every human has a mother, humanity does not. (Our summary.)

Where the debate stands (our summary)Avicenna argues that a set of caused things requires an outside cause; al-Ghazali judges that this assumes what needs proving. The same quarrel reappears in Hume, Russell and Rowe: does explaining each part explain the whole? An open point (see Leibniz's argument).

Averroes 1180 · attacks premise 3

Avicenna makes existence something added to the essence, like an accident. This is a mistake: "existent" does not name one more quality common to all things, but the fact that the thing truly is what it is. Without this distinction, the idea that an essence "receives" its existence from a cause, and that only God has an essence identical to his existence, loses its support.

a defender replies

No reply possible: objection raised in 1180, after the author's death (1037).

Thomas Aquinas defender of the argument · 1255

Summary: in On Being and Essence (ch. 4), Aquinas takes up Avicenna's distinction between essence and existence. In any being whose essence is not its existence, existence comes from elsewhere; there must therefore be a being whose essence is its existence, which has no other cause. Aquinas follows Avicenna here.

The debate goes on (our summary) Avicenna does hold that existence is other than essence (one can understand what a triangle is and doubt that it exists), and that the First has no essence other than its existence. The question has been debated ever since: Aquinas takes up Avicenna's distinction, others judge it merely mental. Scholars also debate whether Averroes reads Avicenna correctly when he speaks of an "accident". (Our summary.)

Where the debate stands (our summary)Avicenna distinguishes essence and existence; Averroes judges that existence is not a property added to essence. Aquinas takes up Avicenna's distinction, others judge it merely mental. The core disagreement is about the nature of existence and remains debated.

Hume (Cleanthes, in the Dialogues) 1779 · attacks premise 6

Even granting a necessary being, why could it not be the material universe itself? Nothing proves that it is a God distinct from the world.

Avicenne replies

Avicenne

The existence of the necessary existent cannot at all result from the gathering of a multitude: what is necessary in itself has no parts.

The debate goes on (our summary) Written seven centuries before Hume, the thesis answers his question directly: a universe made of parts depends on its parts, so it cannot exist by itself. But Hume adds at once that we do not know all the qualities of matter: unknown qualities might make its non-existence contradictory, exactly as is supposed of God. (Our summary.)

Where the debate stands (our summary)For Avicenna, a universe made of parts depends on its parts and so is not necessary in itself; for Hume, unknown qualities of matter might make its non-existence contradictory. The core disagreement: can one speak of a necessity proper to matter without defining it in advance?

Hume (Cleanthes, in the Dialogues) 1779 · attacks premise 2

"Necessary in itself" makes no sense: whatever we conceive as existing, we can also conceive as not existing. So no being is such that its non-existence is contradictory.

a defender replies

No reply possible: objection raised in 1779, after the author's death (1037).

Bruce Reichenbach defender of the argument · 1972

Summary: discussing another use of the same reasoning by Hume (we can conceive an event without a cause, so it is possible), Reichenbach judges that moving from what we can conceive to what is really possible confuses a condition of our knowledge with a condition of being: one can conceive a necessary truth as false. Applied here, that we can conceive a being as absent does not show that it could really fail to exist.

The debate goes on (our summary) Avicenna could not know this objection, and we have not found a passage that answers it directly. His texts nevertheless offer material for a reply: one can understand what a triangle is and doubt that it exists, because its essence is other than its existence; but "the First has no essence other than its existence" (Metaphysics, VIII, 4). For his defenders, then, the First cannot be conceived as non-existent, if it is conceived correctly. Hume would answer that we have no clear idea of such an essence, and Kant that existence is not a property an essence could contain. (Our summary.)

Where the debate stands (our summary)Hume holds that whatever we conceive as existing we can conceive as not existing; defenders answer that conceivability measures our thought, not being. Avicenna could not know the objection. The core disagreement concerns the link between conceivability and possibility.

Kant 1781 · attacks premise 6

To go from "a necessary being exists" to "it is the supreme being, one and perfect", one must assume that a concept suffices to determine what this being is. That is the very idea of the ontological proof: the cosmological proof is merely the ontological proof in disguise.

no reply

No reply possible: objection raised in 1781, after the author's death (1037).

The debate goes on (our summary) Avicenna's defenders answer that he does not start from the idea of a perfect being: he starts from the fact that something exists, and then derives, from necessity of existence alone, the oneness and simplicity of the First. Kant would answer that this passage from a concept to an existence is precisely what he disputes. (Our summary.)

Where the debate stands (our summary)Avicenna starts from the fact that something exists, not from a concept; Kant judges that the final identification of the necessary being reintroduces a concept. The reply by a merely "factual" necessity, discussed in Aquinas's argument, fits Avicenna poorly, since his necessary existent is necessary by its very essence.

What today's voices say

Defend it

Mohammed HijabHis book The Burhan restates Ibn Sina's argument and says it is the argument most resistant to counterattack. In 2026 he adds that it avoids debates about time, causation and infinity, and that even an infinite series of contingent things stays contingent.

In the same family · Cosmological