David Berlinksi, an agnostic philosopher and mathematician, summarizes one of the philosophical arguments for a past-finite universe as follows:

“If a series of causes do not start, it cannot get going, and if it does not get going, then there will be no intermediate causes, and if there are no intermediate causes, then over here, where we have just noticed that a blow has caused a bruise, there is no explanation for what is before our eyes.[1] Either there is a first cause or there is no cause at all, and since there are causes at work in nature, there must be a first.[2]

Essentially, the argument is that the universe must have a finite past because we experience intermediate causes, and intermediate causes would not be possible if they were not preceded by a first cause that began the whole series.

Berlinksi thinks this is a weak, but not an absurd argument. Personally, I am conflicted about its cogency. On the one hand, it does appear weak. It rests on a tautology, and begs the question. To see how, let’s put the argument in deductive form:

P1 The universe consists of a causal series of events
P2 If a causal series does not begin, it cannot get going
P3 If a causal series cannot get going, there will be no intermediate causes
P4 We experience intermediate causes
P5 Therefore a causal chain got going
P6 Therefore a causal chain began
P7 Therefore the universe began to exist in the finite past

Premise 2 is clearly a tautology. It could be restated as “only that which begins, starts,” or “only that which begins, begins.” Not only is this an unhelpful truism, but it begs the question. In saying a causal series cannot “get going” unless it “begins” is to assume from the start what needs to be proved: that the universe “got going” as opposed to “has always been going.”[3] Says who? If the universe is eternal, nothing ever “got going,” and yet, clearly, there exists a causal series that “is going.” One cannot just stipulate that a causal series cannot exist unless it had a beginning, and then conclude that since a causal series does exist, it must have begun. One must demonstrate why it is that a causal series that did not begin cannot be.[4] I think this can be done.

A causal chain must begin with a first cause to avoid the problem of the impossibility of traversing an infinite. Just as it would be impossible to reach the top step of an infinite staircase, it is impossible to traverse an infinite number of past moments to reach the present. An endless series of events, by definition, has no end, and yet today would mark the end of that series. The concept of an infinite past, then, is incoherent. The past cannot consist of an infinite number of causal events. There had to be a first cause in the finite past that caused all subsequent, intermediate causes. In other words, it is impossible for there to be intermediate causes unless they find their origin in a first cause.

While this vindicates the conclusion of the original argument, it rests on premises that are so different that it can hardly be said to be a modification of the original argument. It seems to be a separate argument altogether. In deductive form the “modified” argument would appear as follows:

P1 The universe consists of a causal series of events
P2 If the causal series is infinite, today would mark the completion of that series
P3 It is impossible to complete an infinite series of events
P4 Therefore the causal series of events had a beginning
P5 Therefore the universe began to exist in the finite past

Does this mean, then, that the original argument should be discarded? Is the tautological nature of premise 2 beyond repair, dooming the entire argument? What do you think of my criticism of the argument? What about my logic? I have been thinking and writing on this for about five hours now. The more I think about it, and the more I write about it, the murkier it gets in my mind. I would appreciate the input of someone who is taking a fresh look at this argument for their insights. Thanks!

[1]Berlinski seems to have made a mental slip in saying the “blow has caused a bruise,” for this assumes a casually-connected chain of at least two events. But according to the argument, no causal relationship can be established between any events unless there is a first cause. As the argument goes, not only would we be unable to explain the cause of the blow, but we would be unable even to say the blow was causally related to the bruise.
[2]David Berlinski, The Devil’s Delusion: Atheism and Its Scientific Pretensions (New York: Crown Forum, 2008), 68-69.
[3]Not only does premise 2 beg the question, but it stacks the deck in favor of a past-finite universe by declaring that a causal series of events, to be an actual causal series of events, must have a beginning event. This is little short of definitional fiat.
[4]It should be pointed out that the defender of an eternal universe has no explanation for the causal series. If only that which has a beginning needs a cause, and the causal series has no beginning because the universe has no beginning, then there can’t be any cause or reason for it. It’s just been going on forever, inexplicably. This is not an intellectually satisfying answer, for it goes against our causal intuitions that everything has a cause, and an infinite regression of causes is impossible. Our causal intuitions tell us the regress must stop at some ultimate cause.