I’m not going to try to fully explain and evaluate Swinburne’s Cosmological and Teleological arguments for God here. That would be way too much to tackle in one or two blog posts. There are just a couple of doubts or concerns about these arguments that I would like to express and explore.
Swinburne’s inductive cosmological argument for God has just one premise:
e. A complex physical universe exists (over a period of time).
g. God exists.
Swinburne argues that e is more likely to be the case if God exists, than if God does NOT exist. From this he concludes that the e represents legitimate inductive evidence for the existence of God; that is to say, the truth of e increases the probability that God exists relative to the a priori probability that God exists, relative to the probability that God exist given only tautological truths (truths of logic and math and analytic conceptual truths) as background knowledge.
If g represents the hypothesis that God exists, and k represents background knowledge consisting only of tautological truths, then Swinburne argues for the following claim:
1. P(e|g & k) > P(e|k)
(Read this as asserting: “The probability of e given g and k is GREATER THAN the probability of e given only k.”)
From premise (1), Swinburne infers the following:
2. P(g|e & k) > P(g|k)
(Read this as asserting: “The probability of g given e and k is GREATER THAN the probability of g given only k.”)
One objection that has been raised against this argument is that it is not clear that a probability can be reasonably or justifiably assigned to a factual hypothesis given background knowledge consisting in only tautological truths. If “The probability of e given only k” cannot be reasonably or justifiably determined (or estimated), then we are in no position assert that some other probability is greater than (or less than, or equivalent to) “The probability of e given only k”.
The same issue arises with claim (2) that Swinburne infers from claim (1). If “The probability of g given only k” cannot be reasonably or justifiably determined, then we are in no position to assert that some other probability is greater than (or less than, or equivalent to) “The probability of g given only k”.
But this issue with the idea of a probability given only background knowledge consisting of tautological truths is not the concern I wish to explore here. My concern is with the other conditional probabilities in these equations:
P(e| g & k)
P(g| e & k)
I’m not sure that these probabilities make sense either. My concern is this: Is it possible to know just one contingent fact? Is it possible to know that ‘God exists’ without knowing any other contingent facts? Is it possible to know that ‘A complex physical universe exists (for a period of time)’ without knowing any other contingent facts? If it is not possible to know just one contingent fact, or if it is not possible to know only the contingent fact that ‘God exists’ or to know only the contingent fact that ‘A complex physical universe exists (for a period of time)’, then it appears that we are being asked to conceive of a set of circumstances that is logically impossible.
If it is not possible for a human being to know just one contingent fact, these expressions might still be meaningful and useful as abstractions, as tools of hypothetical reasoning. Arguments typically have just a few premises, and we evaluate arguments by focusing in on these questions: Are each of the premises clear and unambiguous? Are each of the premises true? If all of the premises were true, would the conclusion follow logically? or would the conclusion be made probable assuming the premises were true? Does any of the premises beg the question at issue?
However, if knowing that g is true requires that one knows some other things as well, if knowing g presupposes knowing q, then objections to the knowability of q also work as objections to the knowability of g. So, the epistemological presuppositions of knowing g or of knowing e are relevant to evaluating Swinburne’s cosmological argument.
Suppose I know the fact that I am 5 feet 8 inches tall. Suppose I know that ‘Brad Bowen is 5 feet 8 inches tall’. Can I know just this contingent fact and no other contingent facts? Let’s think about this for a bit. I must understand that the name ‘Brad Bowen’ refers to a specific person, a specific human being, and that the measurement here relates to the size of the human body that belongs to a specific human being. I suppose that all of this could be taken as conceptual knowledge, as knowledge involved in simply understanding the meanings of the words and phrases in the sentence ‘Brad Bowen is 5 feet 8 inches tall’.
To have a clear and correct understanding of this sentence, I must also know that while many animals walk on four legs, human beings walk on two legs and use their arms for other purposes. Thus, the height of a human being is not measured when the person is on his or her hands and knees. Also, height at least for human beings, is measured when the person is standing, not when the person is horizontal, as when the person is sleeping. I should also know that rulers or yardsticks or measuring tapes are used for measuring the height of humans. This assumes that there are physical substances that are fairly stable in their length. Rulers and yardsticks don’t generally grow or shrink large amounts in short periods of time. A ruler that is 12 inches this morning is not likely to be 24 inches this evening. A yardstick that is 36 inches today is not likely to be 25 feet tomorrow. Furthermore, human height is significant and relevant in part because it is relatively stable, at least for periods of days and weeks. I was only about two feet tall when I was born, was about four feet tall when in elementary school, and was over five feet tall in high school. People usually get taller rapidly as young children and teenagers, and then their growth in height slows, and height is stable for many years.
As you can see, there is a fair amount of background knowledge involved in knowing the fact that ‘Brad Bowen is 5 feet 8 inches tall’. Some of that knowledge is conceptual/linguistic knowledge, but some of the knowledge mentioned above is contingent factual knowledge about the world and about human beings. If such an apparently simple and innocuous fact as this requires a good deal of background knowledge in order to clearly and fully understand and know the fact to be true, then I suspect that a deep philosophical claim like ‘God exists’ or ‘A complex physical universe exists’ also requires a significant amount of background knowledge to clearly and fully understand that claim.
To be continued…