05 / 06

Reflections on "Uncaused Beginnings"

William Lane Craig

Used by permission of Faith and Philosophy 27 (2010):  72-78.


Graham Oppy’s interesting analysis of the “causal shape” of reality conflates causal ordering with temporal ordering of causes and assigns the wrong causal shape to reality as conceived by many classical theists. His argument for the possibility of uncaused beginnings is also hobbled by his tendency to ignore the crucial issue of the objective reality of tense and temporal becoming. Oppy’s claims that only certain types of things can come into being uncaused at a first moment of time and that things cannot now come into being uncaused are examined and found implausible and explanatorily vacuous.


Although Graham Oppy’s interest in the possibility of uncaused beginnings springs from his concern with cosmological arguments for God’s existence, [1] the truth of the causal premiss featured in at least one version of the argument, namely, that everything that begins to exist has a cause, is of such general metaphysical importance that it ought to interest any metaphysician. Unfortunately, Oppy’s article gets off to an uncertain start as a result of certain problematic features of his characterization of fundamental notions in his introductory section.

First, there is the ambiguity of what is meant by an “initial state.” Later sections of the article make it clear that Oppy’s concern is with temporally initial states. But that is not how initial states are characterized in his introductory section. Rather, there states are repeatedly said to be ordered “under the causal relation.” Thus, the series of states so ordered could all be simultaneous. States which form a circle under the causal relation, for example, need not require that time is cyclical, for the states may all obtain at once, rather like the four-intercalated flaps of a box top, each holding down another. Thomas Aquinas’ argument against an infinite regress of causes, which plays so central a role in the first three of his Five Ways, similarly concerned causes which are, as he put it, “essentially ordered,” rather than temporally or “accidentally” ordered. On Aquinas’ view, it is a matter of indifference whether temporal states take the accidental causal shape of Regress, Circle, or Contingent Initial State—in each case states must be sustained in being by God, who is prior to all other states under the ancestral of the essential causal relation. In order to rule out such possibilities, Oppy must be assuming that causal directionality entails temporal ordering of cause and effect. But the metaphysician will rightly be sceptical of any such attempt to rule out the possibility of simultaneous, essentially ordered causes by mere stipulation.

Second, even if we suppose that states which are ordered under the causal relation are necessarily also ordered under the earlier than relation, it is unclear why “naturalism would be preferable to theism,” as Oppy claims, if states of reality had the causal shape of Contingent Initial State. [2] Indeed, a good many, if not most, contemporary theistic philosophers hold that this is the causal shape of reality. For the initial state prior to all others under the ancestral of the causal relation is taken to be God’s bringing the universe into being. Since God’s so doing is a free action, such a state is contingent, despite the metaphysical necessity of God’s existing. Reality thus has the causal shape of Contingent Initial State. Mutakallim, or proponents of the kalam cosmological argument, given their strong commitment to divine freedom, [3] embrace this view, rejecting Regress, Circle, and Necessary Initial State as possible causal shapes of reality. Ironically, then, mutakallim will reject the first premiss of the argument Oppy reconstructs, namely,

1. If it is possible for reality to have a contingent initial state under the causal relation, then it is possible for other (non-overlapping) parts of reality to have no cause.

Proponents of the kalam cosmological argument will maintain that while reality must have a contingent initial state if the existence of the universe is to be plausibly explained, it is impossible for any existing thing, whether occupying an initial state or a later state of reality, to come into being without a cause.

Third, Oppy’s reconstruction of the argument on behalf of the causal premiss of the kalam cosmological argument neglects a crucial assumption of kalam: the view that time is tensed and temporal becoming is an objective feature of reality. Oppy’s tenselessly formulated premisses are entirely compatible with a metaphysic of tenseless time, according to which the parts of reality referred to in the premisses do not come into being at all but simply exist tenselessly at their appointed stations. On such a tenseless view of time, it is far less obvious that parts of reality which exist later than the initial state must have causes and that the things existing tenselessly at the initial state cannot exist without a cause, since they do not come into being at that time. If Oppy is to allow the argument in question its full intuitive force, then it must be reformulated along tensed lines. For example:

1´. If it is possible for something to come into being without a cause at a first moment of time, then it is possible for things to come into being without a cause at later moments of time.

2´. It is not possible for things to come into being without a cause at later moments of time.

3´. Therefore, it is not possible for something to come into being without a cause at a first moment of time.

The driving force behind the arguments is the intuition that there is nothing about temporal moments as such that could make their location relevant to whether something can spring into being at that point without a cause of any sort.




In section 1 of his article Oppy discusses how a naturalist would defend the claim that while it is possible for reality to have an uncaused, contingent initial state, it is nonetheless impossible for later parts of reality to be uncaused. To my mind the argument of this section constitutes a virtual reductio of the naturalist’s position. [4] As Oppy explains, the naturalist will say that the contingent things that feature in the initial state of reality are the only kinds of things that can have no cause. Since we are discussing naturalism here, the claim is effectively that, for example, only some sort of fundamental entity described by a quantum gravitational Theory of Everything can pop into being uncaused but that later things, say, tigers or bottles of root beer or Beethoven must have causes of their coming into being. As Oppy explains, this claim plausibly commits the naturalist to two further claims: first, that such entities can come into existence only at a first moment of time and, second, that things which come into being at later times could not come into being at a first moment of time. As Oppy recognizes, the naturalist is thereby committed to holding that the entities in question have the essential property of coming to exist only at a first moment of time and other things all have the essential property of existing only at embedded moments of time.

These suppositions strike me as simply fantastic. Why is it that only a certain kind of particle, say, can come into being uncaused at the first moment of time? Obviously we cannot say that nothingness has a peculiar disposition to producing such particles, as so saying reifies nothingness and invests it with properties. It is baffling, then, to see why, say, a bottle of root beer might not have featured in the initial state of reality. (If one stumbles at design features of a bottle of root beer, substitute other elementary particles like photons or electrons.) Equally baffling is the claim that it is impossible that particles which sprang into being uncaused at the first moment of time should do so later on. Since no causes govern their coming into being, one would think when they do so is wholly arbitrary.

Oppy’s “clear answer” to these questions is merely a reiteration of the naturalist’s commitments. It is explanatorily vacuous. What we want to know is why the entities have these odd essential properties, which are, after all, not qualities of the entities in question but more like arbitrarily asserted predications masquerading as qualities.




In section 2 Oppy takes cognizance of the objection that it is arbitrary for the naturalist to single out certain sorts of things as uncaused features of an initial contingent state of reality, since there is nothing to constrain what comes into being at that moment. Oppy rejoins that if it were really possible for a rabbit, say, to feature in an initial state of reality, then it would have to be possible for the initial state to be nothing but a rabbit, which is manifestly impossible. Of course, I agree that it is obviously impossible that the first state of reality should be a rabbit, since I hold that things cannot come into being uncaused. But, first, I fail to see why it is impossible, if things can come into being without a cause, the initial state might not have been a single rabbit. It might exist for no more than an arbitrarily short moment, but that is irrelevant. A fundamental failing of Oppy’s argument in this section seems to be his assumption that the things which feature in an initial state of reality must be capable of persisting, which is just not true. Besides, we can substitute other entities which can exist apart from larger networks, such as various elementary particles. If quantum gravitational particles can feature in the initial state of reality, why not electrons instead? Second, even if rabbits can exist only as parts of larger networks of related entities, we are still left wondering why, if things can really pop into being without a cause, whole interrelated networks do not and cannot pop into being without a cause. Why could our whole solar system not pop into being uncaused with uncaused rabbits scampering about on an uncaused Earth? We can make the network as large and elaborate as desired; after all, nothing constrains us.




In section 3 Oppy tackles the deeper metaphysical claims about modality and causation that would render the naturalist’s position plausible. But it seems to me that the account of modality and causation proffered by Oppy in order to explain how only the things featured in an initial contingent state of the universe can come into being uncaused is explanatorily vacuous, being tantamount to mere reassertion of his afore-stated naturalistic principles. Its stipulation that all possible worlds have the same initial state as the actual world is an ad hoc conjecture no more plausible than the naturalistic principles limned in section 1. So I do not see how these views about modality and causation can be said to provide serious support for Oppy’s principles.




In section 4 Oppy focuses on the universe as we know it in order to construct a plausible case for the claim that things cannot now come into being uncaused, a case which is compatible with things’ coming into being uncaused at the beginning of the universe. The rough idea is that before some concrete object can occupy the space currently occupied by another concrete object, the current occupant must vacate the space to make room for the new object. The “before” here has to do with priority “in the causal order,” not the temporal order. In the temporal order the evacuation and occupation of the space are presumably simultaneous. According to Oppy, it cannot be that the coming into existence of the new object brings it about that the former occupant ceases to occupy the space, since non-existent things do not have causal powers and the new object does not begin to exist until it occupies some space. Therefore, Oppy infers that the former occupant’s ceasing to occupy the space is a cause of the new object’s coming into being. Therefore, the new object does not come into being uncaused after all.

This account of the matter strikes me as perverse. Just as a bubble forming in a liquid displaces some of the liquid occupying a certain spatial volume, so an object coming into being without a cause displaces any object currently occupying a certain spatial volume. If this is correct, then the new object’s occupying the region is causally prior to the former occupant’s vacating it. Oppy thinks that this account cannot be correct because the new object does not begin to exist until it occupies some spatial location. Now the “until” here must have reference to the temporal order of events, and certainly the new object does not exist before it occupies a certain volume of space. In the temporal order, its existing, its occupying a certain spatial volume, and its displacing the former occupant of that volume are all simultaneous. But such coincidence in the temporal order has no relevance to the causal order. In the causal order the new object’s coming into being at some place causes the former occupant to vacate the space. At the time this happens the new object does exist; indeed, this event occurs at the first moment of its existence. By contrast, on Oppy’s view the current occupant’s vacating the space causes the new object to come into being, which is clearly wrong-headed. Indeed, on Oppy’s view we are left wondering why an object of precisely the same shape and size did not come into being as a result of the evacuation of a certain spatial region by an object. Why would the movement of a table cause a tiger to pop into being?

What is especially strange about Oppy’s line of argument is that his disallowing things to pop into existence uncaused at non-initial moments of time is based, not on the intuitively obvious truth that things like tigers cannot pop into being, but rather on the alleged fact that their popping into being is not, strictly speaking, uncaused. Tigers and the like on the proffered account can, indeed, pop suddenly into existence when something moves out of their way. The only difference between them and the quantum gravitational particles which pop into being at the beginning of the universe is that space comes into being along with the particles rather than precedes them. This is hardly a realistic and convincing account of why, if such particles could pop into being at the first moment of the universe’s existence, they do not do so now nor why other things are not popping into existence now. Oppy insists that his account does not stipulate that the sole cause of things’ popping into being is the existence of appropriately shaped spaces which are consistent with their occupation by an object of a certain kind. Granted, but we still are left wondering why such events are either impossible or have a vanishingly small probability.




In the final section of his paper, Oppy recognizes that his theorizing may have gone wrong in various ways, but he claims to have achieved the main aim of the paper, namely, to show (i) that it is possible for naturalists to engage in this kind of metaphysical theorizing, and (ii) that there can be no justified assessment of the theoretical merits of the products of such theorizing that does not look at the details of the theories in question. That this is the paper’s main aim comes as a bit of a surprise, since that aim is considerably more modest than the aim staked out in the paper’s initial abstract, namely, to defend the view that it is possible for reality to have an uncaused initial state even though this is impossible for any later state. That being said, I hardly imagine that anyone should care to deny (i). But (ii) is importantly ambiguous. Obviously, one cannot assess the theoretical merits of the products of someone’s theorizing unless one knows what those specific products are! But (ii) is, I fear, an expression of Oppy’s general strategy to forestall indefinitely acceptance of cosmological arguments of natural theology by so overloading their proponents with a multitude of profound philosophical conundrums before such arguments can be considered to be good arguments that a kind of paralysis is induced. [5] This sort of suspension of judgement with respect to the causal principle that everything that begins to exist has a cause has not been justified, I think, by Oppy’s paper. That principle is so perspicuous that it is reasonable to rest with one’s affirmation of its truth unless and until naturalistic theorizing such as Oppy envisions serves to dislodge it from our convictions. Oppy’s disquisitions in this paper hardly rise to that level.

Talbot School of Theology

  • [1]

    See his Philosophical Perspectives on Infinity (Cambridge: Cambridge University Press, 2006), ix, as well as his Arguing about Gods (Cambridge: Cambridge University Press, 2006), 148-53.

  • [2]

    It is worth noting that even Regress need not imply that physical states of reality are sempiternal. So thinking implicitly presupposes that the causal states so ordered are all isochronous, i.e., of the same temporal duration. But if the states are allowed to progressively diminish in duration as one regresses in time, then the regress may be of finite duration even though it lacks an initial state. Even if there was no first instant of the universe’s existence, the universe may nonetheless be finite in the past and therefore, in that sense at least, have begun to exist, thus requiring a transcendent cause. Reality as a whole, however, would not be characterized by Regress because the state of God’s causing the universe would be an initial state of reality. If we deny that causal priority implies temporal priority, then there is still a causally initial state of reality, viz., God’s bringing the universe into being, even if there is no initial temporal state of reality, so that the correct model would be, not Regress, but Contingent Initial State.

  • [3]

    As comes to expression in their espousal of the Principle of Determination, on which see my The Kalam Cosmological Argument (London: Macmillan, 1979), 47-49, 150-51.

  • [4]

    Oppy’s analysis is thus helpful in exposing the implausibility of Wes Morriston’s claim that while things cannot come into being uncaused at embedded moments of time, they can do so at a first moment of time (Wes Morriston, “Must the Beginning of the Universe Have a Personal Cause?” Faith and Philosophy 19 (2002), 94-105). As Oppy shows, Morriston is thereby plausibly committed to Oppy’s implausible theses that certain things are essentially capable of coming to exist only at a first moment of time and other things essentially capable of coming to exist only at embedded moments of time.

  • [5]

    See Oppy, Arguing about Gods, 170-71 and my complaints about this strategy in my critical notice of Arguing about Gods, by Graham Oppy, Philosophia Christi 10 (2008), 435-42.