Within any given period of history, the particular inquiries of men and women are, in large part, determined by the dominant values that pervade the age. For the medieval thinker nothing could have been of greater value than the knowledge of God. Theology was esteemed by most as “queen of the sciences.” – a title not merely indicative of the high value placed upon theology in the middle ages, but also, and more importantly, one which is illustrative of the epistemic status enjoyed by the discipline at that time. Theology was seen as consisting in a kind of knowledge. One which was of no less epistemic value than that of philosophy. And it was undoubtedly the project of natural theology – that is, the branch of philosophy which seeks to demonstrate the existence and attributes of God through reason alone – which was, at least to a large extent, responsible for cultivating such an attitude.
Enter William of Ockham (ca. 1285-1349). Though, Ockham himself was a man enamored of both philosophy and (sacred, or revealed) theology, part of his work it seems can clearly be read as an earnest, and for the most part, effective effort to keep the two spheres distinct (Kaye par. 93). Philosophy for Ockham is utterly incapable of offering any conclusive support to theological principles (including the existence of God), which, as Ockham maintained, could only be known by faith (Baird and Kaufmann 357). From even a cursory glance at the secondary literature it becomes clear that, insofar as objections to the conclusiveness of the various proofs of God’s existence are concerned, Ockham should be seen as specifically taking issue with the philosophy of John Duns Scotus (Leff 382). Indeed, in his Quaestiones, the entire section devoted to the issue of whether or not God’s existence can be demonstrated philosophically centers around a kind of dialogue with Scotus’ arguments (Boehner 115-22). Nevertheless, it is my view that in arguing against the efficacy of philosophy to arrive at a knowledge of God, he was arguing for a dissolution of the “great medieval synthesis,” and as such was involved in an indictment against all of those philosopher-theologians who had come before him, and who had sought to demonstrate the reasonableness of a natural theology by means of so-called proofs of God’s existence (hereafter: Proofs). This means that Ockham raises opposition, not merely to Scotus’s arguments alone, but to the Proofs in general. And it is this relationship – specifically that of Ockham’s refutation of the possibility of a Proof from causality to the cosmological argument(s) extant at that time – with which I will be concerned.
In this essay I’d like to focus on one of the more philosophically interesting arguments that Ockham levels against the pursuits of the natural theology of his day. First, I will briefly illustrate the cosmological argument from causality, paraphrasing a form of the argument contemporary with Ockham. Then I will consider the objection that Ockham himself raises against such a line of reasoning. Elucidating his argument’s implications, and revealing what I take to be a major flaw in it’s crucial premise, I hope to demonstrate why Ockham’s objection is ultimately unsound. Thus the reader will be left to draw his or her own judgments concerning the conclusiveness of the cosmological argument in light of the present considerations.
The cosmological argument is really a category of related arguments, all of which seek to demonstrate the existence of a first cause (i.e. an uncaused/necessarily existent cause), which in turn is identified as God. Thus a cosmological argument takes the shape of an argument from universal causality (Cosmological par. 1). By the time of William of Ockham, many different versions of the argument had been developed by a range of philosophers including Plato, Aristotle, Ibn Sina, al-Ghazali, Maimonides, Anselm, and, yes, Aquinas and Scotus (Moreland and Craig 465). However, “the” cosmological argument may roughly be rendered as follows:
Everything which begins to exist has an (efficient) cause
nothing which begins to exist can be it’s own efficient cause.
A causal chain cannot be infinite in length.
Therefore, there must exist a first (or uncaused) cause (Cosmological par. 9).
According to Sharon Kaye, Ockham maintained that the most plausible versions of either the ontological or cosmological Proofs could be reduced to an infinite regress argument taking the form: (1) “If God does not exist, then there is an infinite regress” (2) “Infinite regresses are impossible” (3) “Therefore, God must exist” (par. 81). As Kaye mentions however, though Ockham accepts the impossibility of the kind of infinite regress implied in the second premise of this argument, he nevertheless rejects the premise as inconclusive. If we are to understand Ockham at this point, says Kaye, it is important to understand the distinction that Aristotle made between an extensive and an intensive infinity (par. 83). An extensive infinity is “an uncountable quantity of actually existing things” (par. 83). Ockham recognizes this concept as logically contradictory, because if the objects actually exist, they would in fact constitute an actual number, which, no matter how astronomical, would in principle be countable (i.e. the quantity would really be finite) (Kaye par. 83). An intensive infinity, on the other hand, is something which is “infinite” by virtue of the fact that it has no limitation. The set of whole numbers, for instance, would constitute an intensive infinity, because there is no limit to how far “up” one may count. However, as Kaye points out, “This does not mean that the set of whole numbers is an uncountable quantity of actually existing things” (par. 83).
Kaye goes on to consider the implications of this for the cosmological argument in particular. First, “According to Ockham,” she says, “advocates of the cosmological argument reason as follows: There would be an infinite regress among causes if there were not a first cause; therefore there must be a first cause…” (par. 88) The conclusion clearly hinges upon the second premise of the infinite regress argument mentioned above – namely the impossibility of an (extensive) infinite regress – again, a premise with which Ockham fully agrees. In the Quaestiones Ockham considers “whether a first efficient cause can be sufficiently proved by production as opposed to conservation.” (Boehner 118). By “production” and “conservation,” Ockham means efficient causality, and conserving causality respectively. An efficient cause in this sense would be a cause that brings something about, or which produces an effect “successively over time” (e.g. my grandparents are the efficient cause of my parents, who are the efficient cause of me etc.) (Kaye par. 89). A conserving cause is a cause which maintains the existence of an effect; or put differently it is a cause who’s own continued existence is the necessary condition for the continued existence of it’s effect (e.g. oxygen in the room is the conserving cause of the candle’s flame) (Kaye par. 89). Ockham denies that the First Cause can be conclusively demonstrated on the basis of efficient causation (Boehner 120). Furthermore, though he believes that the inference to a First Cause could be made on the basis of conserving causation, he apparently rejects this, too, on the grounds that this type of causation cannot be demonstrated. Again, to quote Kaye:
In Ockham’s view, the cosmological argument fails using either type of causality. Consider efficient causality first. If the chain of efficient causes that have produced the world as we know it today had no beginning, then it would form, not an extensive infinity, but an intensive infinity, which is harmless. Since the links in the chain would not all exist at the same time, they would not constitute an uncountable quantity of actually existing things. Rather, they would simply imply that the universe is an eternal cycle of unlimited or perpetual motion. Ockham explicitly affirms that it is possible that the world had no beginning, as Aristotle maintained.
Next, consider conserving causality…if the world has to be “held up” by conserving causes, then there must be a first among them because otherwise the set of conserving causes would constitute an uncountable quantity of actually existing things (this is because a conserving cause must continue in existence, simultaneously with it’s effect, in order for it’s effect to continue in existence -AGA)… [Ockham] points out, however, that, just as the cosmos need not have a beginning; it need not be “held up” in this way at all. Each existing thing may be its own conserving cause. Hence the cosmological argument is entirely inconclusive. (par 90-92)
In other words, of the two distinct types of causation, it is only efficient causation which is necessary and demonstrable. Unfortunately for the cosmological argument, a chain of efficient causes does not necessarily imply an extensive (i.e. contradictory) infinity. Ockham maintains, albeit hypothetically, that this causal regress could very well constitute a mere intensive infinity. Thus one would be able to continue back through the causal chain without limit. Like the set of whole numbers there would be no end point. As Ockham puts it, “We can go on with causes of the same kind ad infinitum” (Boehner 119). Notice that the crucial premise upon which the conclusion rests is that it is possible for an infinite regress of efficient causes to constitute an intensive infinity. If it could be shown that it is not possible that an infinite regress of efficient causes could constitute an intensive infinity, then Ockham’s objection that the cosmological argument is inconclusive would fail. And for this reason, I think it does.
Consider first what is meant by an infinite regress. This would be a succession of backward movements ad infinitum. Now recall that by intensive infinity we mean a kind of potential infinite, i.e. much like the set of whole numbers, an intensive infinity is a collection, whose infinity is constituted by successive addition without limit (like counting from zero up the number line). In order for an infinite regress to be intensive, it would clearly have to constitute a backwards-moving succession, whose backwards movements could be “added to” without limit (like counting from zero back through the whole set of negative numbers). But does this make any sense when we apply it to an actual sequence of past events? It would appear not.
First, it should be made clear that the past, because it constitutes a collection of events/entities which have actually already taken place, is by it’s nature definite and determined. By definite and determined I mean something which is of necessity fixed or static. Anything which has the potential to be altered in actuality is ipso facto not definite or determined. It follows from this (by modus tollens) that since the past sequence of events is determined, then the past does not have the potential to be altered in any way. Such alterations would include of course addition to, or subtraction from, or any change in the order of past events. In short, the past cannot be changed.
Now at this point one might object on the grounds that we are, in fact with every passing moment, adding to the past. This however would not be an addition in the same sense as what is meant above. When we consider the forward motion of time, yes, there is a sense in which we are “adding” to the past; but this isn’t addition in the sense of altering what was already there. The type of addition which is precluded by the past’s fixity is the addition of an event which would somehow actually alter the past. The “addition” of new events into the past through the natural passing of time does not alter past states of affairs, and therefore does not contradict the past’s determinacy. In fact, once a present state of affairs becomes part of the past it too becomes fixed and unalterable.
So what type of addition to the past do I mean, which is precluded by the fixity of the past? This would be any addition – of an event or an entity – which would somehow entail an alteration to the past state of affairs. This could be, for instance, “adding” an event prior to some other event which had already occurred, for this would certainly entail an alteration to the past state of affairs, and would therefore be impossible. Nevertheless, this type of addition to the past of prior events is precisely what is necessarily entailed in the concept of an intensive infinite regress of past causes. Given what we have seen of the concept of an intensive infinity, it follows that an intensive infinite regress of causes would constitute a collection of temporally sequenced causes extending into the past, to which prior causes could be added without limit. An intensive infinite regress of causes requires the past to have the potential to be altered in actuality. In fact, the regress of past causes in an intensive infinity can only be infinite in the sense that one can continue on adding prior causes ad infinitum. But this is impossible. For it involves the logical contradiction of not only somehow altering a fixed state of affairs, but also of some event/entity existing prior to the time at which it began to exist.
It could be objected that this is a mere misunderstanding of our concepts or terminology. One could say that it is only a certain sense in which an intensive infinite regress of causes “adds” prior events to the past. The key to avoiding the absurdity is in understanding that they are prior events, and as such were already there the whole time, prior to the moment which we thought we were “addinig” them. I don’t object to this. Unfortunately this would mean that the chain of past temporal causes would not in fact constitute an intensive infinity, but rather an extensive infinity. “If there has been a sequence composed of an infinite number of events stretching back into the past, then an actually infinite number of events have occurred” (Moreland and Craig 473). Thus Ockham, would have been wrong in claiming that the regress of causes could possibly constitute an intensive infinity. This is because as we have seen, there can be no potential causes in the past, only actual causes. But if there can be no potential past causes, then the regress of past causes cannot in principle constitute an intensive infinity; If there can only be actual past causes then this would entail that, if infinite, then the sequence of past causes must be an extensive infinity. But this too leads to logical contradictions which Ockham, no less than his predecessors, understood. This of course means that there is no sense in which a past chain of actual causes (whether efficient or conserving) could be infinite without necessarily running into a contradiction. It would seem then, that the crucial premise of the cosmological argument from causality escapes the razor unscathed.
Works Cited:
Baird, E. Forrest, and Walter Kaufmann. Philosophic Classics: From Plato to Derrida. Upper Saddle River, New Jersey: Pearson Education, Inc., 2008.
Boehner, Philotheus. Ockham – Philosophical Writings: A Selection. Indianapolis Indiana: Hackett Publishing Company, 1990.
“Cosmological Argument.” Wikipedia. January 21, 2010: 12 secs. January 26, 2010.
Kaye, Sharon. “William of Ockham (Occam, c.1280-c.1349).” Internet Encyclopedia of Philosophy. January 3, 2007: 10 secs. January 25, 2010.
Leff, Gordon. William of Ockham: The Metamorphosis of Scholastic Discourse. Oxford Road, Manchester: Manchester University Press, 1975.
Moreland, James Porter, and William Lane Craig. Philosophical Foundations for a Christian Worldview. Downer’s Grove, Illinois: InterVarsity Press, 2003.