Monday, September 5, 2011

An Argument from Contingency

1. There exists at least one contingent being.

2. There is no possible world in which only contingent beings exist.

3. Therefore, in a possible world in which one contingent being exists there exists a necessary being.

4. Therefore, a necessary being exists.

(1) is indisputable for any serious thinker. The justification for (3) is both an analysis of (2) and (1) (since there is no possible world in which only contingent beings exist and this possible world contains a contingent being) and because the only way of denying (3) while granting (2) is to assert that it is impossible that contingent beings exist, denying (1) (meaning that either none of us exist or we’re all necessary beings). Of course, then the conclusion follows.

What may be controversial is (2). The atheist or skeptic may wish to deny this premise in order to avoid the conclusion.[1] It seems we can imagine a world of contingent beings only. But wait—on what are they contingent ultimately, if not a necessarily-existent being? The only answer I can seem to provide is that the existence of reality is just a metaphysical brute fact of reality. That is to say, there just is no explanation of anyone’s existence ultimately. This seems to be a conversation stopper more than anything else.

Now I do recognize that most atheists will not be convinced by this reasoning, quite possibly because it is so abstract. They also may not be convinced because (2) is the operative premise. (1) is accepted by most people, and if people accept it and (2), then (3) and (4) follow necessarily. It is important to understand what the denial of (2) means; it means there is a possible world containing only contingent beings. But since what it means to be a necessary being is occupying all possible worlds, a denial of (2) entails that there are no necessary beings.

One may also object that one will not believe (2) without already believing the conclusion. However, I think this avoids question-begging because of our a priori (at least to this argument) understanding of what it means to be a contingent being; it derives its existence from something else. Because of that, one may well be comfortable in granting (2) without already believing that a necessary being exists. That an acceptance of (2) entails the conclusion of the argument is simply an a fortiori experience for these people.

                [1] They will want to deny the conclusion, for it is an unacceptable response to claim the universe may be that necessarily-existing being. Most people are unwilling to accept the universe is an actual being (save for pantheists).

All posts, and the blog Possible Worlds, are the sole intellectual property of Randy Everist. One may reprint part or all of this post so long as: a) full attribution is given (Randy Everist, Possible Worlds), b) all use is non-commercial, and c) one is in compliance with the Creative Commons license at the bottom on the main page of this blog.


  1. Interesting thoughts. I think you were right to be concerned about the similarity of (2) and (4); I was wondering what was in one that was not in the other.

  2. Thanks Tom. Well, strictly speaking, from (2) it only follows that (3) is true. (4) could be false if (1) were to be denied. However, to me this just shifts the concern of question-begging from (4) to (3), and doesn't get me out of the water yet!

    This was mostly a "trying out just to see" kind of thing, and not meant to be wildly successful. Interesting, to be sure. I do think it avoids question-begging so long as the average person doesn't rely on (3) or (4) in asserting (2). But they may not, even if (2) entails (3) analytically (suppose they have not realized this, or consider [2] apart from [3] and find it plausible that because of contingent beings' dependence on other things, some necessary being exists. That it entails the conclusion is not itself indicative of question-begging). That said, I don't think it will be convincing to anyone who does not already accept the conclusion. So, oddly, I think it is not necessarily question-begging while acknowledging its functional question-begging. :)

  3. Hi Randy,

    This of course depends on a few things. First, it depends on whether an infinite regress is strictly possible, as we discussed a few weeks ago. You could have an unbroken causal chain stretching into an infinite past. It's controversial, but I don't htink it's conclusive.

    Second, and probably the more persuausive point, is to question the reasoning specifically about "being." If being is something personal, then I do think we can deny (2) quite easily. We may instead say that some sort of space-time is necessary and from that beings may arise in some possible worlds.

    Because of counter possibilities, I don't think a deductive case will work here as well as some sort of inductive or best explanation case.

  4. Hi Dr. Mike. I'm mostly in agreement. I don't think there's anything compelling for an agnostic or atheist here. I do like the argument, but I find it's because I believe it is sound already. I don't think being comes from non-being, and I don't think an infinite regress is non-vicious in these cases, so I find (2) not only plausible, but probably true. But most atheists, upon reflection, think being not only can come from non-being, but probably does; and there's the hang up. :)

  5. Right. I do think it probably does. In fact, there is some instance of it we see everyday - ourselves! This isn't a very satisfying example, but it's an interesting one. In some sense, we are all built out of non-being. We consist of quarks and all the same building blocks of everything else. I don't have any good answer for what produces the elan vitale, but it seems clear that the components for being and non-being are the same.

  6. Just a minor point regarding (2):

    I'm not quite sure why we should regard an entity's modal status as a contingent or necessary being as having much to do with its status as an ontologically free-standing being. That is, why does a contingent being have to be an ontologically dependent being? I'm thinking along the lines of exapologist here:

  7. Hi Jake. I'll have to read the article later, but what would the alternative be in regards to contingent objects not deriving their existence from something else ontologically? Brute facts? I hesitate to embrace that account if that is the case.

  8. Hey Randy,

    Ex has a number of posts relating to these kinds of Leibnizian concerns, and I would recommend his posts generally, if you should have time to check it out.

    Anyway, the idea is that there are four logical possibilities concerning metaphysical modality and ontological dependence. Let N be modal necessary and let I be ontological independence. For a given entity E, we have the four logical possibilities: N^I, N^~I, ~N^I, and ~N^~I.

    In the first category we have, for instance, the purported God being and some conceptions of abstracta. In the second category, we have some other conceptions of abstracta (as being dependent upon God) as well as certain accounts of the divine trinity (where the Son is necessary yet ontologically dependent upon the Father). In the last category, we have common things like cars and tables and chairs.

    The question, then, is why we should say the third category is empty. After all, all the above argument from contingency gets us to, strictly speaking, is an ontologically independent being, not a modally necessary one. What would be absurd, for instance, in an essentially independent, contingent being? This is, after all, the kind of being that I take theistic philosopher Richard Swinburne to understand God to be. I think many atheists would be inclined to see the universe this way also. Personally, I am rather undecided on this point.

    Would the existence of a contingent, independent being be objectionable as a "brute fact"? Well, I don't see why it would be any more of a brute fact than the purported metaphysically necessary beings. It looks as if the objectionableness of brute facts would have to do with tracking the chain of ontological dependence. But, ex hypothesi, these beings would be every bit as ontologically independent as the necessary independent beings, so should be no more objectionable on those grounds. And, further, it's questionable just how much explanatory power the consideration of an independent being as necessary would have over considering it as contingent. If one were to hold to the view that even the fact of this contingent, independent being's existence needed to be explained, then we run into van Inwagen's BCF critique of these sorts of factual PSR's.

    So what we need is an *argument* for why modality for contingent beings must perfectly coincide with ontological dependence.

  9. Hi Jake. I think the main thrust of the issue is that something's just existing with no explanation whatsoever (a non-necessary but not-contingent) seems to be counterintuitive at the very least. You're right that Swinburne thinks (or at least he did in the 70s anyway) that God is a contingent being, but that is clearly in the minority amongst theistic philosophers.

    It seems to me the comparison to metaphysically necessary beings or things is solved when considering that a necessary being must exist! If the causal or dependence chain is reasoned backward to a beginning (or even infinitely further back, if one wishes) with no ultimate explanation, then there just is no reason something exists rather than nothing, and these unexplained contingent facts have no explanation of their lack of explanation; something I do find metaphysically absurd!

    I have come to realize this type of an argument relies on the PSR, so that the Leibnizian Cosmological Argument or argument from contingency is better suited for discussion. :)


Please remember to see the comment guidelines if you are unfamiliar with them. God bless and thanks for dropping by!