The Cosmological Argument
The cosmological argument is very old and tries to prove the existence of God as the cause of the universe. It comes in four main varieties:
1. The argument that the series of contingent beings that makes up the universe requires a necessary external cause, identified with God.
2. The argument that since the universe had a beginning and what begins to exist has a cause, the universe has a cause, identified with God.
3. The argument that the existence of the universe is more likely given God’s existence than given the lack of God’s existence.
4. The argument that since the universe exists contingently it must be kept in existence (caused to exist) by a necessary being, identified with God.
Here we look only at the first three. For a brief account and critique of the fourth, see the article by Gale in C 86-101, especially 90-91.
THE COSMOLOGICAL ARGUMENT-(1)
The best version of (1) was provided by Samuel Clarke, a contemporary of Newton.
1) Nothing cannot cause anything.
2) Something exists now.
3) There is a cause or reason for everything
NOTE: this is called the Principle of Sufficient Reason
4) Hence, at no time there was nothing, i.e., something has been in existence
from eternity.
5) This something is either:
a) an independent being, i.e., a being which has
within itself the reason of its own existence (i.e., God)
or
b) an infinite chain of dependent beings.
Now Clarke tries by a reductio to show that (5b) is false.
6) Suppose (5b) is the case.
7) since each member of the infinite chain is a dependent being, it follows
that no cause can be given for the existence of the series itself since:
a) nothing exists outside the infinite series to cause
it)
and
b) nothing in the infinite series is the cause of the
series but at most a cause only of what comes after it.
8) Hence, the series exists without cause or reason, which is impossible.
9) Hence (5b) is false.
10) Hence, (5a) must be the case, i.e., there exists from eternity an
independent being, i.e. God.
NOTES:
· The argument moves within Newtonian cosmology, where space and time are infinite, continuous, and necessarily existent.
· The divine being exists necessarily but is not self-caused, Clarke claims.
· The argument does not intend to prove that the world has a beginning: not having a beginning is one thing, existing necessarily is another. For Clarke, as for Aquinas, scripture, not philosophy or science, tells us that the world had a beginning.
PROBLEMS:
· Is Clarke justified in considering (5a)-(5b) as exhaustive alternatives? He adopts a discrete picture of the universe’s existence, but what about invoking a continuous or even a dense series and ending up with a half open (time-0, now] interval to model the temporal existence of the universe?
· The series over and above its members does not exist in reality; it is just a concept because it is nothing but a set: its components are not parts but elements. Hence, when each member is accounted for, no further question can be asked, since a set is completely determined by its members. So, Clarke has no reason to look for the cause of the infinite series each of whose members is accounted for.
· Do we know that PSR is true? Couldn’t there be fundamental brute facts? Certainly, the standard interpretation of QM suggests that there are. PSR has pragmatic justification, but the relevant use in the argument is epistemic, not pragmatic.
· How do we know that the necessary cause is the God of any particular religion, or even a sentient being?
THE COSMOLOGICAL ARGUMENT –(2)
It is often referred to as the Kalam argument from the Islamic school of thought that first proposed a version of it. It amounts to the following:
1. Whatever begins to exists has a cause
2. The universe began to exist
3. Hence, the universe has a cause, God
Problems:
· (1) is a casual version of PSR, and some, e.g., Hawking, deny it on quantum-mechanical grounds, although the primordial vacuum out of which the quantum fluctuation resulting in our universe allegedly came was not an Epicurean vacuum, i.e., nothing. Moreover, although at present the prevalence of evidence seems against it, one might adopt a big-bang/big-crunch account, as Wheeler did in the 1970’s. Our physical understanding of these issues is far from settled, and therefore philosophical arguments based on it are more than usually speculative..
· How do we know that the necessary cause is the God of any particular religion, or even a sentient being?
· Quentin Smith has argued that since the existence of the universe at time-0 is a physical impossibility, as it is a singularity with the four-dimensional universe occupying a non-dimensional point, and the four-dimensional manifold of general relativity is continuous, one must think of the existence of the universe as involving a continuous open interval as the time variable tends to zero. But then the existence of the universe at any time is explained by that of the universe at a previous time. This makes a good case for atheism as it nullifies the causal role of God.
THE COSMOLOGICAL ARGUMENT –(3)
While the two previous versions of the argument are deductive, Swinburne has proposed an inductive one. He distinguishes between two types of inductive arguments. Suppose that P is the set of premises and C the conclusion. Then
· P-inductive arguments make their conclusion more likely than its negation: Pr(C|P) > Pr(~C|P)
· C-inductive arguments, which merely claim that the premises increase the probability of the conclusion: Pr(C|P) > Pr(C)
Swinburne presents what he takes to be many C-inductive arguments which cumulatively establish that the existence of the universe makes God’s existence more likely than not.
Let G= “god exists” and U=”The universe exists” and let Pr(A|B) mean ‘the probability of A given B.’ In probability calculus,
a. Pr(G|U) > Pr(G)
if and only if
b. Pr(U|G) > Pr(U).
Intuitively, this makes sense, as we can see if we set G=”the wife did it” and U=”the wife’s fingerprints are on the murder weapon”. So, if the physical universe is more likely to exist given that god exists then the existence of the universe provides evidence for the existence of god. Swinburne claims that we should accept (b) is true because although Pr(U|G) may be low (god does not need the world) Pr(U) is still lower because
· The appeal to god is the ontologically simpler hypothesis, as god is a simple being, while the universe is a composite
· The appeal to God solves the causal regress problem (god, as a necessarily existent being, is uncaused) thus eliminating the appeal to brute facts
So, although the existence of the world does not make god’s existence very probable, it still increases it a bit. Since the same type or argument can be repeated with respect to other features of the world such as its order, intelligibility, the existence of persons (the sort of things a personal god would presumably produce), the cumulative evidence provided by all such C-inductive arguments amounts to a P-inductive argument in which Pr(G) > ½.
Problems:
· It’s unclear whether the god hypothesis is ontologically simpler than its rivals such as quantum fluctuations, black holes generating universes, etc., as it involves the positing of a unique, supernatural, necessarily existent entity with infinite degrees of positive qualities. To say that such a being is simple requires fancy theology, such as demanding that all the divine attributes are really one; things get even worse if one thinks about the Trinity and Christology.
· The god hypothesis is not explanatorily simple. For example, if the Christian god made the world, then we got the problem of evil; we have no real sense of god’s intentions, plans, etc.
· Nobody knows what Pr(G) is. If this prior probability is very low, then Pr(G|U) may still be very low. (The rare disease example)
· Is this god the god of any specific religion?