Saturday, April 4, 2015

A Cumulative Case for God's Existence

Richard Swinburne is famous for his inductive arguments which lead to a cumulative case for God's existence.  He distinguishes a correct-C inductive argument (in which the evidence X increases for Y more than in the absence of X) from a correct P-inductive argument (in which case, the overall balance of the evidence for Y is more probable than not).  Let's consider some inductive arguments, even if they are not Swinburne's.

The Cosmological Argument

1. Complex things most likely have an external cause. (Premise)

2. The universe is complex. (Premise)

3. Therefore, the universe most likely has an external cause. (From 1 and 2)

The Teleological Argument

1. Whatever lacks intelligence and exhibits regularity is most likely the result of design. (Premise)

2. The laws of nature lack intelligence and exhibit regularity. (Premise)

3. Therefore, the laws of nature are most likely the result of design. (From 1 and 2)

The Moral Argument

1. If people have knowledge of objective moral obligations, then they most likely have an intrinsic end designed for such knowledge. (Premise)

2. People have knowledge of objective moral obligations. (Premise)

3. Therefore, people mostly likely have an intrinsic end for knowledge of objective moral obligations. (From 1 and 2)

The Argument from Happiness

1. People who pray to God and meditate on average live longer, healthier, and happier lives than those who don't, as long as it is not at the expense of others. (Premise)

2. Self-interest is worth pursuing so long as it is not at the expense of others. (Premise)

3. Therefore, one should most likely pray to God and meditate. (From 1 and 2)

The Argument from Desire

1. Innate desires most likely correspond to something that can satisfy them. (Premise)

2. Perfect and eternal happiness is an innate desire. (Premise)

3. Therefore, one's innate desire for perfect and eternal happiness most likely corresponds to something. (From 1 and 2)

The Argument from Religious Experience

1. Barring some defeater, one's religious experience is most likely genuine. (Premise)

2. There are those who have religious experiences that lack any defeater. (Premise)

3. Therefore, there are those whose religious experiences are most likely genuine. (From 1 and 2)

The Argument from Suffering

1. A perfect God would not allow gratuitous suffering. (Premise)

2. Most likely, whether there is gratuitous suffering is unknowable. (Premise)

3. Therefore, it is unknown if there is gratuitous suffering and if God allows it. (From 1 and 2)

Now, let's weigh the evidence.  Surely there are other arguments, but let's stick with these for now.  The arguments in favor of God's existence end up like this: P(h/e&k) > P(h/k).  In English, P = a proposition's probability, h = a hypothesis, e = evidence, and k = one's background knowledge.  We can know this is the case, since these are all facts we would expect if God exists, rather than if he didn't.  The argument from suffering fails as a correct C-inductive argument because it is inscrutable.  There is no way of determining whether gratuitous suffering actually occurs.

If all of this is correct, then we actually have a correct P-inductive argument: P(h/En&k) > 1/2.


  1. The problem with this is that it is also possible to make a cumulative argument against the existence of God.
    So, you have to take into account the arguments against God too and only then can you assess the overall probaility.
    Another problem here is that only one of your arguments that is actually an argument for God and that's the teleological argument. Even that argument doesn't even mention personality.
    The cosmological argument argues for a cause beyond the universe and doesn't even mention intelligence and neither does the argument from desire.
    The moral argument is just about the weakest argument ever, since it is based on some very vague inituition and no theist can even account for objective morality under God. BTW, Swinburne himself doesn't think the moral argument is sound.

    If you weigh that against only 2 atheistic arguments , like e.g; the argument against the possibility of creatio ex nihilo, or even the evidential problem of evil, you don't get your 1/2 by a long shot.
    To summarize, while it is possible in principle to build a cumulative case for God, so far nobody has sucessfully done this. But who knows?

    1. If you want to make a cumulative case for atheism, be my guest. However, Swinburne has already gone over them and considers only one to be a correct C-inductive argument.

      I think you're also overlooking that this is a cumulative argument. You're right that the cosmological argument I've presented doesn't give us God's personality but, combined with the other arguments we gain knowledge God's additional attributes. It's like traveling through Europe. We travel to one part and found rivers; another part and you find mountains, and so forth. As a result, you're able to describe Europe as having all of these different attributes. The same is true with a cumulative case for God.

      Moreover, Swinburne accepts the version of the moral argument I've laid out. He's talking about human knowledge of objective moral obligations, not what grounds morality. According to him, morality is independent of God, something that distinguishes him from classical theists. I suggest reading these arguments more carefully.

    2. You are correct about Swinburne's moral argument, but I am not overlooking that it is a cumulative argument. You cannot get to God's 'attribute' of personality from a cumulation that doesn't mention personality. Each of God's 'attributes' needs a cumulative argument and your version doesn't succeed in doing that.

      It is also a fact that there are philosophers who have gone over each of Swinburne's arguments and some of them don't consider any of then to be a correct C-inductive argument for God. So, that's a matter of opinion.

  2. Another problem is that postulating an entity that can do everything will in many cases automatically makes X more probable than in the absense of such entity. The real question is: how likely is such entity?
    Consider a very good stage magician. He pulls rabbits out of a hat, flies around, cuts woman in two, disappears and reappears somewhere else etc.
    Is this more likely if he has true magic powers than if he has to rely on tricks? Of course.
    Do we therefore conclude, after seeig all his performances, that the probability of him having true magic powers is > 1/2? No, we don't, because the prior probbability of him having those powers is extremely low.

    1. In this case, P(e&k/h) < or = P(k/h) because the evidence is consistent with one's background knowledge that these are just tricks. These feats don't make it more likely that he has genuine magical powers for the simple reason of our background knowledge.

    2. I am not taking into account our background knowlegde that these are just tricks. I am talking about the prior probability of having genuine magical powers. For someone who believes in magic, this probability is higher than for someone who doesn't. Likewise, the probability of God having the powers required for this cumulative case to work is,to an atheist, extremely low. A theist will of course disagree.
      So, as I said, the real question is not whether the probability of X is greater if a personal omnipotent being who wants X and is able to produce X out of nothing exists than if such being doesn't exist.
      The real question is: is it probable that such being exists?

  3. I think that formulation of the CA is flawed, or at least could stand to be re-elaborated. While it makes sense to talk about objects in the universe as having an external cause, singe they're spatiotemporally located, it doesn't seem that we can extend the same principle immediately to the universe as a whole, since speaking of the universe as spatiotemporally located is just circular; spatiotemporal locations are a feature of being "in the universe". (I think we may have had this conversation before.)

    Also, have a happy Easter.

    1. Happy Easter to you, as well. We could reformulate the argument if you wish. I wrote this in a hurry. However, do you really think the universe as a whole isn't complex? Taken as a whole, I would think it's the most complex thing in reality.

  4. First of all, everyone, thank you for your comments. I truly appreciate the feedback, whether critical or affirmative. For awhile I'm only going to reply to comments on occasion. You can blame graduate school and having to write my thesis on that. So, see you soon. I'm not going anywhere.

  5. Replies
    1. From one Dutchman to another, thank you. :)

    2. Since I am actually Belgian, that would be a no true Dutchman fallacy, I guess.

    3. I'm part Belgian too, so I recognize your country's independence.