My Formulation of Avicenna's Contingency Argument

By Mohamad Farhat

An examination of Avicenna's argument for the existence of a Necessary Existent.

My Formulation of Avicenna's Argument from Contingency

My version of Avicenna's argument from contingency is based primarily on the fact that the argument works regardless of whether the universe has a beginning. That is, even if the universe is eternal, Avicenna claims that it still needs to be explained. His argument is not focused on finding the "first" moment in time; instead, he asks why there are any contingent beings at all.

In The Metaphysics of the Healing, Avicenna answers this question using the distinction between necessary and contingent existence. According to Avicenna, any being that is contingent needs a cause. He goes on to argue that an order made up of only contingent existents cannot ultimately account for its own existence. I will outline my version of the argument using three premises and a conclusion. Then, I will describe the reasoning behind each of those premises, along with what I believe is its greatest objection.

Argument Outline

First Premise: Each and every existent is either necessary through itself or contingent through itself.

Second Premise: Any existent that is contingent through itself requires a cause to explain why it exists.

Third Premise: An order composed of only contingent existents cannot ultimately account for its own existence, regardless of whether its members constitute an infinite causal regress.

Conclusion: Therefore, at least one Necessary Existent exists.

Contingent and Necessary Existence

In Book I, Chapter 6 of The Metaphysics of the Healing, Avicenna contrasts what is necessary in itself with what is possible in itself. A Necessary Existent is an entity whose nonexistence is impossible. It exists through itself and requires nothing outside of itself for its existence. On the other hand, a contingent existent is an entity whose essence necessitates neither its existence nor its nonexistence.

Take a tree, for example. Nothing in the essence of a tree necessitates its existence. Instead, it depends on several factors, including the seed that produces it, the environmental conditions under which it grows, and the processes responsible for its development. Depending on how those factors unfold, the tree may never have existed. Therefore, the existence of a tree is contingent rather than necessary.

However, one might ask whether a tree that has existed forever would be necessary. Not according to Avicenna. Temporal duration and ontological dependence are two fundamentally different concepts. Something can exist forever and yet depend on some other factor for its existence. Avicenna believed in the eternity of the universe, so his argument cannot be reduced to simply showing that there was a first moment of creation.

Therefore, the relevant question is not how long something has existed, but whether it exists through necessity. This leads us directly into our first premise. Each and every existent, considered in itself, either exists through necessity or does not. If it does not, then it is contingent.

Why Contingent Existents Need Causes

Our second premise follows from the very definition of contingency. If an existent is contingent through itself, then what explains the fact that it currently exists?

Avicenna argues that anything possible through itself requires a cause for the simple reason that nothing about the essence of an existent which exists contingently provides an explanation for why that existent presently exists. Consider a tree once again. We can give reasons for why the tree exists today. These include things like seeds, soil, sunlight, and other environmental conditions. These explanations do not demonstrate, however, that trees exist necessarily through themselves. They simply show what they depend on to exist.

There is a major distinction to draw here. Avicenna believes that something can be necessary through another while existing contingently through itself. So long as the complete cause for an existent which is contingent through itself is in place, that existent's existence is necessitated by its cause. The fact that its existence is necessitated by that cause does not mean, however, that it exists necessarily through itself. Its existence remains dependent on something else.

The same applies to our example of a tree. Its existence can be explained by reference to the conditions that led to its production. Those explanations do not, however, remove its contingency. If A depends on B for its existence, then A remains contingent through itself even when B is identified as its cause. And if B is itself contingent, then B requires a cause.

How far does this series extend?

The Problem of Infinity

Assume we have an infinite causal series in which A depends on B, B depends on C, and so forth without end. Each member of the series has a cause. What is wrong with assuming that we need no more than this? After all, if every existent has an explanation for its existence, what else remains unexplained?

I believe that the source of the problem is the difference between providing explanations for individual members of a causal series and providing an explanation for why the causal series exists at all.

Let us use an analogy to illustrate this point. Imagine an infinitely long chain where every link depends on another link. We could explain what supports any given link by reference to the link immediately above it. However, none of these links provides independent support. Every link remains dependent, and adding an infinite number of additional links does not affect that dependency.

This analogy does not prove our conclusion. It merely illustrates the problem. The real question is whether a series composed solely of dependent existents can provide an ultimate explanation for its own existence.

In Book VIII, Chapter 1 of The Metaphysics of the Healing, Avicenna argues against an infinite regress of essentially ordered efficient causes. Rather than saying that an infinite number of prior events cannot occur, Avicenna argues that an order of causes in which each member derives its causal efficacy from another cannot ultimately be self-sufficient without a cause that does not derive its causal efficacy from another member of that order.

At this juncture, our third premise comes into play. If every member of a causal order is contingent through itself, then every member derives its existence from something else. No matter how far back you trace the causal order, nothing within it has existence necessarily through itself. Although an infinite regress extends the series of explanations, it introduces no member that possesses existence independently.

Objection to the Third Premise

Someone might argue that explaining every single member in an infinite causal regress would suffice to explain the whole series. Since each member has a cause, perhaps explaining all the members means there is no need to look for a further cause for the whole order.

This objection raises a legitimate challenge. Just because each member in a series possesses a particular quality does not imply that the totality possesses that same quality. To assume otherwise without further justification risks committing the fallacy of composition. Specifically, this objection raises the question of whether the contingency of individual members necessarily establishes the contingency of their totality.

However, I am convinced this objection fails to defeat our argument. The central issue here is not simply whether the properties of individual members must also belong to their totality. Rather, the focus is on whether each member depends on others within that totality to explain its existence.

To illustrate, suppose the totality of all contingent existents exists independently of anything beyond it. Thus, it is independent of any external cause. Nonetheless, no member within it possesses independent existence, and each member's causal explanation relies on something dependent that is also part of what requires explanation.

Therefore, if there is a cause for the totality, that cause cannot be one of its members, since a member cannot ultimately account for the existence of the collective to which it belongs without participating in what requires explanation.

An opponent could continue to deny that any cause exists for the totality whatsoever. Perhaps contingent reality simply exists as an inexplicable brute fact. While this represents a possible rebuttal to our argument, it denies that contingent reality as a whole requires an ultimate explanation.

Ultimately, I view this position as weaker than our argument since it leaves an entirely dependent order without any ultimate explanation for its existence.

In summary, I do not view the issue as whether or not infinite chains of dependents are possible. I contend, rather, that the central question is whether an infinite number of dependents constitutes enough to provide an ultimate explanation for their totality.

Establishing the Necessary Existent

Finally, we can establish our conclusion.

For purposes of argument, let us assume that no Necessary Existent exists. By virtue of our first premise, everything therefore exists contingently through itself. Based on our second premise, every contingent existent has a cause for its existence. Reality would thus comprise only entities whose existence depends on something else.

Our third premise establishes, however, that an entirely contingent causal order cannot ultimately account for its own existence. Whether this causal order consists of a finite or infinite number of members matters little. If each member in it depends on another for its existence, then none possesses existence necessarily through itself.

Since this conflicts with our third premise, our original assumption that no Necessary Existent exists must be false.

Thus, we infer that at least one existent exists necessarily through itself rather than through another. This is what Avicenna refers to as the Necessary Existent, or wājib al-wujūd.