In the course of debating theists, there are a variety of arguments for theism that constantly recur. In my experience, many of the people who present these arguments are under the impression that they are either new or irrefutable. Therefore, it may surprise them to learn that all the most commonly heard arguments for God’s existence have been around in one form or another literally for centuries, and all of them were refuted soon after they were first proposed. The only genuinely new argument for the existence of God to emerge within the last hundred years is the presuppositional argument, and even that fails to stand up to scrutiny.
While the failure of these arguments is not, in and of itself, the reason why I am an atheist (see “All Possible Worlds“, “A Ghost in the Machine“, “One More Burning Bush” and “The Cosmic Shell Game” for insight into the reasons for that), it is obviously necessary to show that there is no convincing proof of God’s existence if one is to be an intellectually honest and consistent atheist. This essay will accomplish that, by examining the four classic types of arguments for God’s existence as formulated by theists both historical and modern and demonstrating that all of them are fundamentally flawed in some way. It will also examine some more specific or less frequently heard pro-theistic arguments and show that each of them suffers from irreparable defects as well.
- The Ontological Argument
- The Cosmological Argument
- The Design Argument
- The Moral Argument
- The Presuppositional Argument
- Other Arguments
“Thus even the fool is compelled to grant that something greater than which cannot be thought exists in thought, because he understands what he hears, and whatever is understood exists in thought. And certainly that greater than which cannot be understood cannot exist only in thought, for if it exists only in thought it could also be thought of as existing in reality as well, which is greater. If, therefore, that than which greater cannot be thought exists in thought alone, then that than which greater cannot be thought turns out to be that than which something greater actually can be thought, but that is obviously impossible. Therefore something than which greater cannot be thought undoubtedly exists both in thought and in reality…. And you, Lord God, are this being.”
—Anselm of Canterbury, Proslogion, 1077 CE
The first, and one of the oldest (dating back almost a millennium), of the “classical” pro-theism arguments is the ontological argument. This argument attempts to prove God’s existence from reason alone, without any reference to the external world, based on nothing but the standard definition of who and what God is. In addition to Anselm’s formulation, versions of the ontological argument have been defended by philosophers such as René Descartes, Gottfried Liebniz, Kurt Gödel and Alvin Plantinga (whose formulation of it is considered in the next section). In its classic form, as presented by Anselm, the argument runs as follows:
- By definition, God is the greatest being that can be conceived of.
- But if God did not exist in reality, we could imagine a being that had all the other properties of God but that also existed in reality, and this being would be greater than God.
- Since God is the greatest being that can be conceived of, this is impossible.
- Therefore, God must exist in reality.
While the casual reader may well intuit that this cannot possibly be right, it is easier to say that this argument looks wrong than to say precisely what is wrong with it. It is this property of the ontological argument that has fascinated philosophers for centuries. However, a more detailed analysis will reveal the flaw it contains. In fact, the ontological argument is subtly circular: it “smuggles in” a crucial presupposition. This may not be obvious in the above formulation of the argument, but we can reformulate it to make it more obvious. Consider the following modified version of the ontological argument that makes this hidden assumption clear:
- By definition, God is an omnipotent, omniscient, maximally benevolent being that exists.
- Therefore, God exists.
The fallacy in this should be clear: one cannot simply define something into existence. A person can tack phrases like “and that exists” onto any definition to their heart’s content, but that will not cause that object to exist in reality. Philosophy would be a far more profitable field if this was possible; consider definitions like “the bag full of money sitting next to me right now that exists”. Clearly, this is invalid. A definition, if accurate, truthfully reflects reality; a definition does not create reality. Yet the ontological argument tries to do this very thing, only in a less obvious way.
The key point is this: If it is true that a being that exists in reality is greater than one that exists only in the imagination, then to define God as the greatest possible being is to implicitly claim that God exists – but this is what the argument is supposed to be trying to prove! The ontological argument uses its own conclusion as one of its initial assumptions – it assumes what it is trying to prove, committing the basic fallacy of circular argument. Therefore, it is logically invalid and proves nothing.
Atheists do not accept the ontological commitment that this argument tries to smuggle in through its definitions. It is certainly possible to conceive of a being called God that exists only in thought, and it is also certainly possible to conceive of a being called God that exists both in thought and in reality. One might even be inclined to agree that, if the latter actually did exist in reality, it would be greater than the former. But one cannot validly infer that such a being therefore must exist in reality. Again, this is an attempt to build existence into a definition and conclude that the thing defined therefore exists, and this is fallacious. It is an impermissible attempt to leap up from one level to another. Defining God as a being that must exist can only be accurate if God does in fact exist, and the truth value of that statement can only be ascertained by examining external reality.
“E* entails the property exist in every possible world. E* is exemplified in W*; hence if W* had been actual, E* would have been exemplified by something that existed and exemplified it in every possible world…. That is, there actually exists a being that is omniscient, omnipotent, and morally perfect; and that exists and has these properties in every possible world. This being is God.”
—Alvin Plantinga, The Nature of Necessity, p.215-216; Oxford University Press, 1974.
A somewhat more sophisticated version of the ontological argument has been proposed by the Christian philosopher Alvin Plantinga. Plantinga’s formulation uses the notion of possible worlds; any world that can be imagined to exist, including our own, is a possible world. Plantinga goes on to argue that if there is any possible world in which God exists, which there plainly is, then God must exist in all possible worlds, including this one. In more formal terms, Plantinga’s argument proceeds as follows:
- A being is maximally excellent in a possible world if and only if it is omniscient, omnipotent, and morally perfect in that world.
- A being is maximally great in a possible world if and only if it is maximally excellent in all possible worlds.
- There is a possible world in which there exists a being that possesses maximal greatness.
- Therefore, there exists a being that possesses maximal greatness in every possible world.
- Therefore, God exists.
Though it uses different language, Plantinga’s argument relies on the same fallacious trick of trying to define something into existence. What it essentially says is, “There is a possible world containing an entity that exists in all possible worlds,” or more simply, “By definition, God exists if we can imagine him to exist.” This is not legitimate. Possible worlds, by the definition of what it means to be a possible world, are completely disjoint; they cannot influence each other in any way. (If two possible worlds could affect each other, they would not be separate possible worlds, but the same possible world, again by definition.) Therefore, it is invalid to define something such that its existence in one possible world “causes” it to exist in another possible world. An unlimited array of absurdities would ensue if that was the case.
For a particularly dramatic example of this, consider the following proposition: “There is a possible world containing an entity that destroys all possible worlds.” If we accept Plantinga’s implicit claim that one possible world can be defined in terms of other possible worlds, we must admit that such a world is conceivable. But if there is such a possible world, by definition there will be no possible worlds. Therefore, Plantinga’s logic has led to an obvious absurdity: it has just proved that nothing exists. The solution, again, is to recognize that defining the contents of one possible world in relation to another is logically illegitimate and meaningless.
There is another, very obvious way to counter this argument. Consider this:
- There is a possible world in which there is no entity that possesses maximal greatness.
- Therefore, there is no entity that possesses maximal greatness.
- Therefore, God does not exist.
The logic of this argument would seem to be airtight. Certainly there is no logical contradiction involved in conceiving of a world that possesses no maximally great entity, and therefore this is a possible world; and therefore, by Plantinga’s own logic, there is no such entity in any possible world and therefore there is no God. As yet, proponents of the ontological argument have offered no reply to this.
In this section, I would like to propose one final attack on the ontological argument that, as far as I know, is entirely new. This attack is an example of reductio ad absurdum – disproof by contradiction – and thus, for reasons that will hopefully become clear shortly, I have named it the Manichaean Reductio.
A contemporary of Anselm’s named Gaunilo put forth one of the first and still one of the strongest replies to the ontological argument, in another example of reductio. Using the same logic as Anselm, he proposed the existence of a “perfect island,” one than which no greater can be conceived, and went on to conclude that such an island must exist, because a perfect island that actually existed would be greater than one that only existed in the imagination. And furthermore, we could use the same logic to prove the existence of the greatest possible anything: the greatest possible ham sandwich, the greatest possible ballpoint pen, and so on.
The replies mounted by defenders of the ontological argument have been confused at best. Most of them assert that the analogy is invalid without ever quite explaining why. Some claim that the notion of the “greatest possible” of any object other than God is incoherent and therefore impossible, because no matter how good an island, ham sandwich or ballpoint pen is, we could always conceive of one that was better in some way. (But why does this not apply to God as well?) Although I have never seen any apologist attempt this, one could even theoretically claim that these greatest possible examples of any object really do exist; absurd as this seems, it does not imply any actual contradiction. However, the Manichaean Reductio is a strengthened version of Gaunilo’s attack that not even these weak criticisms can touch. At its core lies a simple realization: why must “greater” be defined in terms of moral goodness?
Suppose we are trying to conceive of the most perfectly good being imaginable, and we reason, as defenders of the ontological argument do, that such a being must exist because one that actually existed would be more perfect than one that did not. But now suppose we try to conceive of the most evil being imaginable. By the logic of the ontological argument, must this being not exist as well? After all, an evil being that existed in reality would surely be superior in evil to one that existed only in the imagination.
By the same logic that leads the traditional ontological argument to conclude that this perfectly good being must be omniscient and omnipotent, we can also conclude that this perfectly evil being must possess these qualities as well. An evil being that was omnipotent would be far more evil than one that was limited, as it would be able to inflict an unlimited amount of pain and suffering on its victims; and an evil being that was omniscient would be more evil than one that was not, as it would be able to know and savor exactly how much suffering it was causing. Therefore, by the logic of the ontological argument, we have “proved” that there is not one but two gods, each one infinite, each one omniscient, each one omnipotent, and the two of them desiring exactly opposite things in any given situation (which follows from the definitions of perfect goodness and perfect evil).
But this is impossible. There cannot be two omnipotent beings that desire different things, because in any given situation, their opposing desires could not both come to pass; the will of at least one would have to be frustrated, leading to the conclusion that at least one was not omnipotent. Yet the ontological argument leads inevitably to this impossible situation. Therefore, since we have used it to derive a contradiction, the ontological argument must be unsound.
Some versions of the ontological argument rely in some way on the claim that God possesses “necessary” existence, meaning it is not possible that God could have failed to exist. But this cannot be true. For something to exist necessarily means that a logical contradiction inevitably follows if one tries to imagine otherwise. (In Plantinga’s formulation, something exists necessarily if it exists in all possible worlds.) It is not possible, for example, to conceive of a triangle with any more or less than three sides, or a prime number that is greater than all other prime numbers, or a married bachelor. However, it is possible to conceive of a world where God does not exist – the existence of atheists proves that. No contradiction is entailed by this proposition. Therefore, God’s existence cannot be necessary.
A theist might object that conceiving of a world without God actually is logically impossible, because God’s existence is required to explain why any world exists at all – but then the ontological argument merely reduces to the cosmological argument, which is dealt with in the next section.
“We see in the world around us that there is an order of efficient causes. Nor is it ever found (in fact it is impossible) that something is its own efficient cause. If it were, it would be prior to itself, which is impossible. Nevertheless, the order of efficient causes cannot proceed to infinity, for in any such order the first is cause of the middle (whether one or many) and the middle of the last. Without the cause, the effect does not follow. Thus, if the first cause did not exist, neither would the middle and last causes in the sequence. If, however, there were an infinite regression of efficient causes, there would be no first efficient cause and therefore no middle causes or final effects, which is obviously not the case. Thus it is necessary to posit some first efficient cause, which everyone calls ‘God.’”
—Thomas Aquinas, Summa Theologiae, c.1260 CE
Dating back to Aristotle, the cosmological argument is very probably the oldest argument offered in support of the existence of God, and probably the most frequently used by lay apologists as well. As formulated above by Thomas Aquinas, the cosmological argument states that every event has a cause; but every cause is itself caused by something else. To avoid an infinite regression, the argument concludes, we must postulate a first cause that is itself uncaused and eternal, and it identifies this first cause as God.
If the flaw of the ontological argument is circularity, the fallacy of the cosmological argument is special pleading. Namely, it asserts without good reason that everything except God needs a cause. But why should this be? If anything can exist without a cause, we could just as well conclude that it is the universe itself that is uncaused, existing eternally and giving rise to all other cause and effect. This hypothesis has just as much explanatory power as the hypothesis that God created the universe, and it is more parsimonious, requiring fewer additional assumptions. Therefore, all other things being equal, it is to be preferred.
Aquinas’ objection to the possibility of an infinite regress is also poorly founded. He claims that an infinite regression of causes could not exist because there would be no first cause, but this shows a failure to understand the notion of an infinite series. In such a series, every individual event would have a perfectly good cause: the event preceding it. Alternatively, if we accept Aquinas’ logic on this point, we can then ask, how many thoughts did God have before creating the universe? Every thought God had must have been caused by another thought preceding it, since Aquinas claims nothing can be its own cause. But since by Aquinas’ argument an infinite beginningless series is impossible, God must have had a single thought preceding all others – i.e., there must have been a point at which God came into existence. We can then ask the cause of this initial thought, and so on ad infinitum.
There is one final attack on the classic cosmological argument. Say for the sake of argument that we ignore the above difficulty and grant this argument everything it asks – then it still does nothing to establish the existence of God. Even if we accept this argument’s logic, all it proves is that there was a first cause. It does not prove that this first cause still exists today; it does not prove that this first cause has any interest in or awareness of human beings; it does not prove that this first cause is omnipotent or omniscient or benevolent. It does not even prove that the first cause is conscious or a person. An atheist could accept this entire chain of logic and then posit that the first cause was a purely natural phenomenon.
“Now from the very nature of the case, as the cause of space and time, this supernatural cause must be an uncaused, changeless, timeless, and immaterial being which created the universe. It must be uncaused because there cannot be an infinite regress of causes. It must be timeless and therefore changeless – at least without the universe – because it created time. Because it also created space, it must transcend space as well and therefore be immaterial, not physical.”
—William Lane Craig, in God? A Debate Between a Christian and an Atheist, p.5; Oxford University Press, 2004.
The modern version of Aquinas’ argument is known as the kalam cosmological argument. More sophisticated than the classic cosmological argument, the kalam argument, as originally developed by Muslim theologians and today employed by Christian apologists such as William Lane Craig, seeks to avoid the charge of special pleading that dogged its predecessor. It most often takes the following form:
- An actual infinite cannot exist in reality.
- Therefore, an infinite number of events cannot have occurred before the present.
- Therefore, the universe began to exist.
- Whatever begins to exist has a cause.
- Therefore, the universe has a cause.
Craig takes this argument further by claiming that space and time came into being along with the universe, and therefore the cause of the universe must transcend both space and time. Finally, and most audaciously, he claims that the cause cannot be an impersonal natural phenomenon, because if the necessary conditions for the universe to exist had existed from eternity, the universe itself would have existed from eternity, and this is not the case; science has discovered that the universe has a finite age, beginning at the Big Bang. Therefore, Craig asserts that only a personal agent could freely will to create an effect in time.
There are several important problems with this argument. The first one lies with Craig’s claimed proof that an actual infinite cannot exist, a claim which he uses to argue that the universe must have had a finite history and therefore a beginning. Craig’s argument for this point relies on alleged self-contradictions that arise when considering the idea of an actually existing infinity. For example, the set of all numbers is infinite in size, as is the set of even numbers, but if we subtract the latter from the former the resulting set is still infinite. More importantly, Craig claims it is impossible to form an infinite set by successive addition – no matter how many times we add 1 + 1 + 1 + 1…, the sum will always be a finite number, never infinity. Therefore, no matter how many past events have occurred, there can only be a finite number of them and there still must have been a first event, a beginning to the universe. While Craig argues that a potential infinite, defined as a value that increases indefinitely without bound, can exist, he denies that an actual infinite can ever exist in reality.
It is true that an actual infinite, if such a thing existed, would possess some very strange and counterintuitive properties; for one thing, such a set could be the same size as one of its proper subsets, which is the source of most of the “absurdities” Craig claims to have pointed out. But this does not prove that such a thing is impossible, merely that the human mind cannot adequately conceive of it. There is no law that requires reality to conform to our expectations. Most people would also find the idea that light can behave both as a particle and as a wave to be counterintuitive or absurd, but nevertheless, quantum mechanics has taught us that it is so.
Regarding the supposed impossibility of forming an infinite by successive addition, Craig’s argument makes a key faulty assumption. Of course an actual infinite cannot be formed by successive addition if one only has a finite number of steps to do it in. But an actual infinite can be formed by an infinite number of successive additions. In other words, there could have been an infinite number of events before now as long as there was also an infinite amount of time before now, which is exactly as we should expect. One might object that this proves that it is necessary to start with an infinite in order to get an infinite. This is true, and it is not a problem if one postulates a universe that has always existed as a brute fact requiring no further explanation, just as theists postulate a God that has always existed as a brute fact.
Finally, there is a problem with this premise that Craig does not seem to have considered, and one that shows why the kalam cosmological argument, despite its greater sophistication, is still built on special pleading. How many things does God know? An omniscient deity, obviously, would know an infinite number of things. How many things can God do? Equally obviously, an omnipotent deity would be able to do an infinite number of things. But these are not potential infinites; they are actual infinites. The number of things God knows or can do, according to traditional theism, is not increasing indefinitely without bound; it is already as great as it will ever be. Therefore, since Craig argues that an actual infinite cannot exist in reality, then he has proven by his own argument that God does not exist – at least, not an infinite god of the type conceived of by so many theists.
We next consider premise 3, which states that the universe began to exist. As Craig puts it, “The astrophysical evidence indicates that the universe began to exist in a great explosion called the ‘Big Bang’ around 15 billion years ago. Physical space and time were created in that event, as well as all the matter and energy in the universe” (p.4). Craig has far overreached himself here. Nothing about Big Bang theory implies or requires that space, time, matter or energy began to exist at that point after previously not existing. Simply stated, the Big Bang was a point at which the universe as we currently observe it was extremely hot and dense. Our current formulations of the laws of physics break down under these conditions, and so we do not know what came before that. But this does not entail that the universe itself came into existence at that point. It might well have existed in another form prior to the Big Bang. It might even have always existed, so that the Big Bang would not be its beginning but merely the least recent event in its history that we can observe. In this case, the kalam argument’s premise 3 fails and therefore the entire argument fails.
It is worth noting that Craig quotes Dr. Stephen Hawking in support of his claim that time itself began at the Big Bang, but takes him out of context in so doing. As Hawking himself has said, even if there were events before the Big Bang, they would make no difference to us. “As far as we are concerned, events before the big bang can have no consequences, so they should not form part of a scientific model of the universe. We should therefore cut them out of the model and say that time had a beginning at the big bang” (Stephen Hawking, A Brief History of Time, p.49; Bantam Books, 1996). In other words, Hawking believes we should say that, as far as we are concerned, the Big Bang is when time began. That does not mean that other events could not have happened before that, merely that it is pointless to speculate because we could never know anything about them. This is not equivalent to Craig’s claim that time did not exist before the Big Bang occurred. Craig is correct to claim that it is the overwhelming scientific consensus that the Big Bang did happen, while alternatives such as the steady-state model are poorly supported by the evidence (p.8); but this does not support his particular interpretation of what the Big Bang may mean for the beginning of the universe. (Craig also erroneously cites cosmological theories such as the oscillating universe or chaotic eternal inflation as if they were alternatives to the Big Bang, when in fact they are attempts to explain what may have happened beforehand. Many of these theories, in opposition to Hawking, do postulate that events prior to the Big Bang may have had observational consequences for the universe).
Next, premise 4 holds that anything that begins to exist has a cause of its existence. But we already know that this is not true. Strange as it seems, science has discovered some natural objects that come into being uncaused. One such class of object is called virtual particles. According to the Heisenberg uncertainty principle, the product of energy and time must always be greater than a certain constant; therefore, as we examine apparently empty space over shorter and shorter time scales, we discover increasingly violent, and increasingly brief, fluctuations of energy. Since energy is equivalent to matter by E = mc2, on time scales that are sufficiently short these fluctuations can be so violent as to create actual particles that exist for only a fleeting instant before returning to nothingness. These particles are not merely a hypothetical construct: they produce detectable effects that can be, and have been, experimentally measured. However, there is literally no specific cause of their existence. Nor is this phenomenon merely an inconsequential anomaly. Some physicists have proposed hypotheses in which a similar phenomenon, under the right circumstances, could have given rise to our universe.
In attempting to refute this point, Craig commits several fallacies. First, he claims that some physicists disagree that quantum fluctuations are uncaused and therefore this is not a “proven exception” (p.6) to this premise. This misses the point. To refute the kalam argument, it is not necessary to prove beyond any doubt that there are exceptions to its premises; such an unrealistically high standard of proof is impossible to meet anyway. It is only necessary to show that such an exception can exist. This would mean that there are alternative explanations for the origin of the universe, and there would therefore be no need to postulate a supernatural creator. In any case, if Craig invokes the support of a solid majority of the scientific community in support of theories such as the Big Bang, he cannot then disregard the weight of scientific opinion when it clashes with another point he wishes to make. Craig cites only one scientist (David Bohm) who believed that quantum fluctuations might be deterministic, and Bohm’s views have not been widely accepted by the physics community.
Secondly, Craig claims that “even on the traditional, indeterministic interpretation, particles do not come into being out of nothing. They arise as spontaneous fluctuations of the energy contained in the sub-atomic vacuum; they do not come from nothing…. Popular magazine articles touting such theories as getting ‘something from nothing’ simply do not understand that the vacuum is not nothing, but is a sea of fluctuating energy endowed with a rich structure and subject to physical laws” (p.6). This, again, misses the point. Craig seems to have forgotten that the objection he is addressing is not whether something can come into being from nothing; it is whether something can come into being uncaused. The two are not the same thing. If I left a pile of lumber and boards in my yard and later returned to find a dining room table sitting there, it would not make the table any less uncaused to say that it was formed from the pre-existing lumber. It is arguably correct to say that virtual particles arise from the quantum vacuum, which is not “nothing”; but in this case one would also have to then say that “nothingness” as traditionally conceived does not exist. In this case, the vacuum would have always existed, and the universe would have come into being from it, quite possibly uncaused.
Finally, even if we overlook the failure of its other premises and assume that Craig’s argument is correct so far – that the universe was caused and an actual infinite cannot exist – this still would not verify the kalam cosmological argument. It still falls victim to the same problem as the classic cosmological argument: even if we grant that the universe had a cause, how do we know the cause is anything at all like the theistic conception of God? How do we know, for example, that the universe was not caused by a timeless, impersonal natural force?
Craig attempts to refute this point as well: “If the cause were a mechanically operating set of necessary and sufficient conditions, then the cause could never exist without the effect. For example, the cause of water’s freezing is the temperature’s being below 0° Centigrade. If the temperature were below 0° from eternity past, then any water that was around would be frozen from eternity. So if the cause is timelessly present, then the effect should be timelessly present as well” (p.5).
Again, Craig attempts to refute the wrong argument here. If the cause of the universe was timeless, it would not have existed “from eternity past”; such temporal language makes no sense if applied to a timeless phenomenon. If time came into being along with the universe, it would make no sense whatsoever to ask why the universe did not come into being at an earlier or a later time. The cause would simply be, and whenever the universe came into existence would mark the beginning of time, t = 0, by definition. If we accept that a timeless cause can produce a temporal effect, there is no problem at all with this. Alternatively, what if the cause is cyclic? It might produce a temporal universe which lasts for a certain amount of time before returning to a timeless state that causes a new universe to come into being. Or what if the cause produces multiple universes, such that our own is merely one in an endless series that extends indefinitely into a timeless past and a timeless future? Craig does not even consider these possibilities, and there seems to be no way to rule them out. The point is that, when entering intellectual realms this vast and uncharted, one must tread all the more carefully, and proponents of the kalam cosmological argument do not do this. As David Hume wrote over 200 years ago in his Dialogues Concerning Natural Religion, “What peculiar privilege has this little agitation of the brain which we call thought, that we must thus make it the model of the whole universe?” Apologists such as Craig would do well to heed such cautions.
Craig says, “The only way for the cause to be timeless and the effect to begin in time is for the cause to be a personal agent who freely chooses to create an effect in time without any prior determining conditions” (p.5-6), but this is just special pleading. On what grounds can he be so sure that an agent can act outside of time while an impersonal cause cannot? We have no experience of any type of cause acting outside time, personal or not, so claims about what types of causes can or cannot act outside of time are very premature and ill-founded at best. It seems that Craig freely exempts his preferred cause from a variety of restrictions which he rather arbitrarily applies to other potential causes.
In summary, the kalam cosmological argument, like its classic cousin, is dogged by the fallacy of special pleading. Its key premises are flawed, and even if we grant them all anyway, its conclusion still does not establish that there is a god, only that there is some cause of the universe’s existence. Whether this cause is personal or not, the cosmological argument cannot tell us, and thus it cannot suffice as a proof of the existence of God.
“In crossing a heath, suppose I pitched my foot against a stone and were asked how the stone came to be there, I might possibly answer that for anything I knew to the contrary it had lain there forever; nor would it, perhaps, be very easy to show the absurdity of this answer. But suppose I had found a watch upon the ground, and it should be inquired how the watch happened to be in that place, I should hardly think of the answer which I had before given, that for anything I knew the watch might have always been there. Yet why should not this answer serve for the watch as well as for the stone? Why is it not as admissible in the second case as in the first? For this reason, and for no other, namely, that when we come to inspect the watch, we perceive – what we could not discover in the stone – that its several parts are framed and put together for a purpose, e.g., that they are so formed and adjusted as to produce motion, and that motion so regulated as to point out the hour of the day; that if the different parts had been differently shaped from what they are, of a different size from what they are, or placed after any other manner or in any other order than that in which they are placed, either no motion at all would have been carried on in the machine, or none which would have answered the use that is now served by it.”
—William Paley, Natural Theology, 1800
The third major class of arguments for God are teleological arguments, also commonly called arguments from design. This argument, which found one of its best-known formulations in the works of William Paley, is an example of argument from analogy. It holds that the products of human design often exhibit multiple parts that work together in a precise fashion to achieve a specific goal, which is an uncontroversial claim. The argument goes on to claim that the world itself also exhibits this property, and so the world must also have been created by an intelligent designer, one as far superior to human beings as the world is superior to the products of human design. This intelligence behind nature, it concludes, can only be God.
The first problem, and it is a serious problem, with Paley’s design argument is as follows. At the conclusion of his argument, Paley claims to have demonstrated the existence of a deity who designed, not just living things, but all that exists. But at the beginning of his argument, quoted above, Paley clearly implies that he recognizes the watch as a designed object because it is in some way different from its surroundings. Why else would he reject the possibility, which he finds entirely reasonable for the stone, that “it had lain there forever”?
And yet, if a pocketwatch – or any other artifact manufactured by humans – looks unlike the objects we find in the world, this is highly damaging to the argument from design, which relies on drawing an explicit analogy between the products of human design and items found in the natural world. The less alike these two categories of things are, the more the argument from design is weakened, and the less likely it is that we can validly infer a designer that is in any way like human beings.
In fact, it is indeed the case that the world looks unlike human design. For example, in the natural world, we see objects that self-assemble from simpler components; that make copies of themselves and repair themselves; that can be grouped into nested hierarchies of classification; and that adapt to previously unencountered situations. These characteristics are generally not found in the products of human design.
Paley’s analogy is weakened by the dissimilarities between human artifacts and natural objects, but it is conclusively defeated by the fact that we now have a plausible explanation, other than intelligent, purposive design, for the complexity we observe in nature. When the teleological argument was first put forth, Charles Darwin had not yet proposed the theory of evolution, but in the hundred and fifty years since that time the theory has attracted a wealth of confirming empirical evidence and the overwhelming support of qualified scientists. Invoking only the two simple principles of random variation and non-random selection, evolution explains how life changes and diversifies over time, resulting in intricate and subtle adaptations to meet almost any conceivable need without requiring an intelligent designer. (Ironically, one of Paley’s chief examples was the eye, the workings of which he explained in detail to prove that it could only have originated through a process of design; and the eye has since become an iconic success story of evolution, as Darwin himself clearly explained the gradations of eyes that exist in nature, from simple to complex, and showed how a selective process could easily have led to the stepwise formation of one from the other.) Similar theories of emergence, self-organization and self-assembly now being studied in many fields likewise explain how complex, patterned behavior can result from the interaction of just a few simple laws, accounting for many more natural phenomena that the theory of evolution does not cover. The advance of science has rendered the design argument obsolete.
One important point is worth making here. Contrary to the unfounded objections of creationists, evolution is not an inherently atheistic theory. It does not, in and of itself, prove or imply that there is no designer. However, it does show that such a being is not necessary in order to account for life’s complexity, which is all that it takes to defeat the design argument.
There are several further objections to the argument from design that, while they do not strictly disprove it, may make it unpalatable to many who would otherwise employ it. For example, why should we infer only one designer? In nature, we observe designs working at opposing purposes – designs to defeat designs. There are pathogenic bacteria with adaptations to defeat the immune system; gazelles with rapid running speed to outrun cheetahs and cheetahs with rapid running speed to catch gazelles; viruses that subvert the natural reproductive machinery of cells to make more copies of themselves until the cells literally burst; and predatory fireflies whose flashes mimic the mating signals of fireflies of a different species to lure their hapless prey, to name a few examples. If we accept the argument from design and then follow it to its logical conclusion, we would have to conclude that there is not just one, but many warring designers. Some theists may have no problem with this conclusion, but many others will doubtless find it unacceptable.
A further implication of the argument from design is that the designer, if indeed there is one, does not seem to care greatly about humans. Paley attempted to infer the designer’s benevolence from the existence of pleasure in the world (going so far as to conclude that swarms of gnats and schools of shrimp seemed to him to be very happy in their day-to-day existence), but any investigation not hobbled by this obvious theological bias would conclude that, in nature, pleasure is in general far outweighed by suffering. All living things must compete fiercely for scarce resources. An enormous variety of infectious diseases and virulent parasites constantly besiege all that lives. Predation is the order of being, as all animals must kill to stay alive. Again, if we follow the logic of the design argument, we would most likely conclude that not only are there multiple designers, most of them are positively malevolent.
One final flaw in the design argument is that it, like the cosmological argument, is built on special pleading. It assumes that the complex and intricate world we live in must be the handiwork of a designer. But surely the mind of any being who could design such a world must be even more complex and intricate than that world is (to say otherwise is to admit that a simpler thing can give rise to a more complex thing, which is highly problematic for design arguments). Who designed God’s mind? If the reply is that God does not need a designer to explain his existence, then why is it that nature does?
“The result of these cumulative efforts to investigate the cell – to investigate life at the molecular level – is a loud, clear, piercing cry of ‘design!’…. Why does the scientific community not greedily embrace its startling discovery? Why is the observation of design handled with intellectual gloves? The dilemma is that while one side of the elephant is labeled intelligent design, the other side might be labeled God.”
—Michael Behe, Darwin’s Black Box, p.232-233; The Free Press, 1996.
Two hundred years later, the design argument has changed very little. The modern version, proposed by biochemist Michael Behe, is almost exactly the same as Paley’s; the only difference is that Behe has conceded that evolution could have created the intricate macroscopic structures Paley so confidently held forth as evidence of design, and instead has shifted his focus to microscopic systems within the cell, molecular relays that perform functions such as allowing bacteria to swim or causing blood to clot. These structures, Behe claims, are too complex for evolution to have produced. Behe calls his criterion “irreducible complexity” – if a system is made up of multiple interacting parts of which all are required to perform its function, then such a system could not have been created through a process of gradual evolution and must have been designed.
One might be forgiven for regarding this as little more than a God of the Gaps argument – when our increasing knowledge means one system is no longer tenable as an example of design, simply move on to another, less well understood, and claim that it is now the newest irrefutable proof of design. There is merit to this charge, but even beyond that the modern design argument is flawed in a more fundamental way. The fact is that evolution can produce irreducibly complex structures, and in fact has been observed to do so, both in living things and in the computer simulations of evolution called genetic algorithms. For examples of this, see this post detailing the evolution of a novel, irreducibly complex multi-step enzyme pathway to break down a toxic man-made chemical, or this article concerning the creation, by a genetic algorithm, of an irreducibly complex circuit composed of 37 logic gates.
How can an evolutionary process give rise to an irreducibly complex system? There are three ways. The first can be summed up as scaffolding: extra parts which support a partially functional system until it is completely assembled, at which point the extra parts become unnecessary and are pruned away by selection. The second, change of function, is a major mechanism of evolutionary change that has been known about since Darwin’s day. A system which originally evolved to perform one function may become co-opted and take on a new function, starting out with multiple functioning parts rather than having to acquire them one piece at a time. Finally, there is the case of improvement becomes necessity, where an adaptation is at first merely beneficial, but as later changes build on it, it becomes indispensable. All three of these mechanisms have been implicated in the origin of various complex molecular systems.
Unlike Paley, who was explicit in claiming that the designer was God, modern advocates of intelligent design are usually coy when asked who exactly they believe the designer to be. (The quoted excerpt above is a rare exception.) This is a legal tactic, not a theological one; they hope to survive the inevitable court challenges to teaching intelligent design in public schools by avoiding explicit public association of the doctrine with religion. Nevertheless, if modern design arguments are to be used as proofs of God, they must face the same theological objections that confront Paley’s argument. Based on what we see in nature, why should we infer one infinite designer, rather than many limited and warring ones? Why should we infer a benevolent designer rather than a malevolent one? (The bacterial flagellum, which has become the icon of the ID movement, is very similar to a toxin-injection system used by several species of bacteria that are infectious and harmful to humans.) And how do design advocates answer the charge of special pleading that stems from their requirement that complex natural systems be designed while the presumably even more complex mind of the designer is freely exempted?
“Therefore, the summum bonum ["greatest good"] is possible in the world only on the supposition of a supreme Being having a causality corresponding to moral character. Now a being that is capable of acting on the conception of laws is an intelligence (a rational being), and the causality of such a being according to this conception of laws is his will; therefore the supreme cause of nature, which must be presupposed as a condition of the summum bonum, is a being which is the cause of nature by intelligence and will, consequently its author, that is God.”
—Immanuel Kant, Critique of Practical Reason, 1788
Moral arguments, the last of the major four classes of pro-theistic argument, take two distinct forms. The first form, as set out in the writings of philosophers such as Immanuel Kant, holds that there is a moral order in the world and that the existence of such a thing can only be explained by supposing the existence of a divine lawgiver. In other apologetic works, this argument frequently takes the form of claiming that any judgment about good and evil presupposes God’s existence because one can only judge the goodness of something by comparing it to God, who is the highest good.
Of course, there are some atheists who would deny that there is any such thing as morality, and while this would defeat the moral argument, I believe that such a victory would come at too high a price. A better solution is to recognize that atheism is fully compatible with a universal, objective moral philosophy, one that is built on reason, empathy and the recognition of our shared humanity. “The Ineffable Carrot and the Infinite Stick” explains this moral system more fully and shows why no god is necessary to account for its existence.
In fact, it could well be argued that atheism is a better foundation of objective morality than theism. The reason for this is that to postulate, as theists do, that a being is the ultimate source of morality suffers from the Euthyphro dilemma: does God approve of something because it is good, or is it good because God approves of it? If the former is the case, then there is an objective standard of morality outside of God, and we can simply bypass God and appeal to this standard directly. But if the latter is the case, then good and evil would be entirely determined by God’s whims, and there would be no genuine objective morality, and thus no moral order, at all. In this respect, the moral argument is self-defeating.
“…we know that men find themselves under a moral law, which they did not make, and cannot quite forget even when they try, and which they know they ought to obey. …If there was a controlling power outside the universe… [t]he only way in which we could expect it to show itself would be inside ourselves as an influence or a command trying to get us to behave in a certain way. And that is just what we do find inside ourselves.”
—C.S. Lewis, Mere Christianity, p.19; The Macmillan Company, 1960.
A slightly different version of the moral argument, advanced by theists such as C.S. Lewis, states that there is a universal moral law, a standard of right and wrong which all human beings are innately aware of, even if some choose to violate it. The argument’s backers claim that such a moral awareness could only have been put into us by God.
The most significant problem with this argument is that human beings are not all aware of the same moral law, as even a cursory examination of human history would reveal. In various societies throughout history, behaviors such as polygamy, segregation, slavery and racism, physical abuse as a method of discipline, infanticide, incest, pedophilia, human sacrifice, ritual suicide, ritual murder, cannibalism, genital mutilation and genocide were widely accepted, even encouraged. None of the societies that did these things seemed to feel that there was anything wrong with them; many justified their actions by appealing to their god. Some societies have shunned violence of any kind, while others have encouraged war and militarism. Some have advocated free speech and individual rights, while others have mandated conformity and the superiority of the state. Even today, there are furious debates over the ethics of topics such as gay marriage, abortion, capital punishment, sex education, drug legalization, contraceptive use and euthanasia, to name just a few. Claiming that God is responsible for humanity’s universal sense of right and wrong fails to explain why there is and has always been such widespread disagreement over morality.
On the other hand, atheism can accommodate both the existence of a moral law and the manifest fact that not every culture or individual is aware of it. The explanation is straightforward: morality is not something implanted in every person’s heart by a creator, but something derived from careful deliberation and a rational understanding of our place in the world and our relationships to each other. There is no reason why we should expect it to be immediately obvious to everyone, just as there is no reason why we should expect the laws of physics to be immediately obvious to everyone.
“The believer in Christ must therefore place himself upon the position of the unbeliever for the sake of argument. The believer can then show the unbeliever that he cannot as much as find a place on which to stand in order to oppose the Christian position. The unbeliever must presuppose the truth of the Christian position in order to deny it. Unless the unbeliever repents and in humble faith accepts the authority of the self-attesting Christ as he speaks in Scripture, he can say nothing about himself as the interpreter of the world or about the world as interpreted by himself. So far as modern thought is not based upon the presupposition of the truth of Christianity it is lost in utter darkness. Christianity is the only alternative to chaos.”
—Cornelius Van Til, A Christian Theory of Knowledge, p.310; Presbyterian and Reformed Publishing Company, 1969.
The final class of arguments for God are presuppositional (also sometimes called transcendental) arguments. This type of argument differs from all others discussed so far in being the closest thing to a genuinely new entry on the list; although it has historical precursors, the fully fledged presuppositional argument was first put forward in the early to mid-20th century by apologists such as Cornelius Van Til and Gordon Clark. This type of argument is also novel in another important way: while the other four classes of argument discussed so far implicitly grant the existence of atheism as a position and then attempt to show why it should be considered false, the presuppositional argument essentially denies that atheism exists as a position at all. Instead, it claims, atheism (or, in fact, any position other than Christianity) is ultimately incoherent, and adherents of such worldviews can only reason or think at all insofar as they borrow the principles that enable them to do it from Christianity. Presuppositional apologists assert that the Christian God is the only possible source or justification for all knowledge about the world, and atheists and others can only come to know anything by implicitly assuming that the Christian God exists, even if they deny that this is what they are doing.
As this essay has endeavored to show, all arguments for the existence of God are ultimately built on fallacies, and the presuppositional argument is no different. Whereas the ontological argument is built on circular reasoning and the cosmological and design arguments on special pleading, the presuppositional argument is built on the fallacy of the false dilemma, which the above selection most clearly brings out. Van Til’s claim is that Christianity – specifically, his brand of Protestant Christianity – is the only alternative to believing that everything is ultimately due to random chance and therefore unknowable. This is absurd on its face. To list a few examples, what would prevent a Jew, a Muslim, a Mormon or a Hindu from using a presuppositional argument of their own and claiming that it is in fact Christianity that assumes the truth of their worldview?
In addition, even if one considers atheism to be the only alternative to Christianity, it is still not true that the choice is between Christianity and chaos. Few, if any, atheists would consider the world as we experience it to be the result of “chaos”; rather, most atheists consider the world to be the emergent result of both regularity and chance, governed by unchanging principles – laws of nature – that we can discover. There is no reason one could not be an atheist and consistently suppose this to be the case, and in fact most do. Van Til’s claim that an atheist must believe that “time and chance… are the ultimate principles of individuation” (p.320) is flatly false, and a brazen straw man to boot.
To finish off the presuppositional argument, this essay will explain in greater detail where the principles of an atheist do come from and why they do not presuppose the existence of a deity of any sort. We will begin with the laws of logic, which are precise descriptions of necessary truths that must hold in any possible world. They are neither arbitrary nor contingent on anything else for their existence. Whether or not a god existed, they would not be any different; nor could a god change them even if one did exist. Similarly, formal fallacies – statements that violate the laws of logic – are necessary falsehoods, statements that are false in all possible worlds. Even God could not make it be Monday and not Monday at the same time; nor does it take a belief in God to understand that proposing “P” and “if P, therefore Q” must necessitate Q.
Next, we turn to the more general matter of how we reason, how we gain knowledge about the world. We learn to reason by observing examples of how the world works and drawing generalizations from these examples which we can apply to other situations in the future. If a generalization fails to hold, we can search for a relevant difference between the situations where it did hold and the ones where it did not to explain this failure, and if we can find no such difference, we may simply conclude that the generalization was incorrect and discard it. No god is needed to explain why this process works as long as there are natural laws that do not change from one situation to another – our generalizations, if correct, will simply be precise descriptions of those laws.
The formalized version of this everyday process is science, and again we find that theists have no advantage here. Science does not presuppose the existence of God – in fact, to the degree it refers to the subject at all, science assumes, if not the nonexistence, then at least the absence of God. This is because science is an inductive process, and induction only works when past experience is applicable to the future – in other words, when there are unvarying natural laws. Induction does not work when we assume that a miracle might happen at any time. This is not to say that a religious believer cannot be a scientist, but it does mean that a religious believer, if they are to do science, cannot assume that God might be secretly intervening at any time to alter natural law and change the results they would otherwise obtain, since they can learn nothing if they make such an assumption. Scientists do not have to be philosophical materialists, but they do have to be methodological materialists. In this respect, neither atheism nor theism is superior in terms of which worldview gives us the ability to gain knowledge about the world.
However, there is another respect in which atheism could be said to better justify the effectiveness of science. A believer who is a scientist could assume that no undetectable miracles occurred while they were doing a given experiment and therefore their results are valid, but why should they assume that? Unless a theist believes in a god whose will is both completely knowable and completely known, they have no grounds on which to assume that he might not completely change the laws of nature at any moment. By contrast, an atheist can consistently believe that the laws of nature are precise descriptions of the ways reality behaves, and cannot be changed since there is nothing “higher” than them that would have the power to do so. In this respect, not only does the presuppositional argument not hold, but the opposite conclusion is true: it is atheism, not theism, that gives us the best expectation of being able to learn true statements about the world.
Finally, we consider morality. Many theists have claimed that we must presuppose God to make any sense of morality, that without belief in an almighty lawgiver there is no reason why human beings should be kind to each other rather than acting on their every selfish wish. However, I claim the opposite is true. Any immoral act can be excused by claiming it is the wish of a higher power that ordered it to be done, and a belief that justice will be served in the end no matter what anyone does gives little incentive for a person to work at improving the human condition. On the other hand, a firm belief that human happiness is the only worthy goal does not allow such easy justifications for evil behavior, and the recognition that there is no compelling evidence for divine justice should lead us to work all the harder because it is up to us. Theism may be compatible with morality, but it is atheism that demands it.
“Jesus made some profound statements about His divinity. We concluded that it is impossible to state He was simply a good moral teacher. From His amazing statements we must conclude Him to be a liar, a lunatic, or God. Since the first two were not conceivable, we began looking at the third alternative. If this is true, we must see if He has the credentials for His claims.”
—Patrick Zukeran, “The Uniqueness of Jesus”, http://www.leaderu.com/orgs/probe/docs/unique.html
In addition to the major classes of arguments discussed above, there are some other, more specific pro-theistic arguments. For example, some Christian apologists defend Christianity through an argument styled the Lord/Liar/Lunatic trilemma. Briefly, this argument states that Jesus claimed to be God, and there are only three possible reasons for this: that he was telling the truth, that he was knowingly lying, or that he was deluded and insane. But, the argument goes, Jesus as depicted in the New Testament gospels clearly did not act like an insane man, whereas his unblemished personal conduct, the greatness of his moral teachings and his refusal to recant even when threatened with death make it implausible to believe that he was lying. The only remaining option, therefore, is that Jesus was telling the truth about who and what he was and thus must have been God.
However, this is a false trilemma. It overlooks a fourth option – lord, liar, lunatic, or legend. How do we know that the gospels are an accurate record of events? What if Jesus’ words and deeds were greatly exaggerated or even outright invented by later writers? (The essay “Choking on the Camel” defends the latter position.) If so, there is no reason to believe Jesus ever claimed to be God at all, or that the gospels’ depiction of his conduct is accurate.
In addition, even the trilemma’s existing options do not capture the full range of possibilities. It is not accurate to say that either Jesus was telling the truth, the whole truth and nothing but the truth, or else he was deliberately and intentionally lying, or else he was ranting-and-raving insane. Many forms of psychological illness manifest themselves in only one way, while in every other respect the individual seems completely sane. Cult leaders, for example, are often very charismatic and persuasive, and are able to attract many followers who do not realize the depth of their delusion. Or perhaps Jesus’ fame as a preacher and faith healer spread so far that he actually convinced himself he was the messiah people told him he was. Although it is a stretch, it is even conceivable that Jesus recognized the cruelty of the Old Testament laws and told a lie for the greater good, deliberately claiming to be the messiah in order to gain the authority to annul them and replace them with better laws, and was so dedicated to this mission that he was willing to sacrifice himself in the end. The point is that the Christian trilemma cannot hold up as long as there are any options other than the three it proposes.
Another argument occasionally encountered by apologists of various faiths is that that faith must be correct because members of it went willingly to their deaths rather than recant. However, this argument is extremely weak. Every faith has its martyrs, from early Christians who were thrown to the lions, to Muslims who blow themselves up out of certainty that they will go directly to Paradise, to Buddhists who set themselves ablaze to protest war and violence. This does not prove anything about which, if any, of these belief systems is correct; it simply proves that every belief system has adherents who are absolutely convinced that what they believe is true. Since most religions are designed to stifle doubt, there is nothing surprising about this. The truth value of a proposition cannot be determined by how fervently people believe in it, but can only be decided by unbiased comparison with the evidence.
Another argument frequently put forward is that we should believe in God because most people throughout history have done so. And this is true, to a first approximation. I freely admit that the vast majority of human beings throughout history have believed in the supernatural.
This is not, however, a valid argument for theism, but a fallacy. If we knew nothing about the world other than that the great majority of people believed a certain proposition, then all else being equal, it would indeed be worthwhile to look into that proposition. (Few atheists would claim that the existence of God is a possibility not even worth considering.) But this would not make that proposition true – that can only be determined by comparing it to the evidence. At one point in history the great majority of educated human beings believed that the Sun revolved around the Earth, but that did not make this claim true. Reality is not determined by majority vote.
A far more important point is that, although most human beings have believed in a god, not all have believed in the same god. Most cultures throughout history have disagreed dramatically, and often bitterly and violently, about the nature and wishes of God. If there truly was a deity giving all people the same information, this would be an extremely strange and unexpected situation; but if there is nothing but humans independently inventing their own ideas, this is exactly what we should expect to find, and so the religious confusion among humanity is not an argument for theism, but for atheism. (“The Argument from Locality” and “The Cosmic Shell Game” develop this line of reasoning further.) Only by ignoring all these stark differences, which touch every single doctrinal point of significance in every religion, can the argument from popularity have any force.
One of the most obvious and most fundamental of these differences regards how many gods exist. Advocates of the argument from popularity tend to belong to the monotheistic Judeo-Christian-Islamic tradition, but most cultures throughout history have not been monotheistic, but polytheistic. If we follow the argument from popularity to its conclusion, we should believe that there is not just one, but many gods.
A very common pro-theistic argument is that God must exist because believers frequently experience a sense of his (or her or its) presence. However, like the argument from popularity discussed above, the argument from religious experience falls to the same counterargument of religious confusion. The claim that God must exist because a particular theist has a sense of his presence does not explain why people claim to have experiences of different gods and supernatural beings. Protestant Christians have religious experiences of a Protestant Christian God; Roman Catholics have experiences of a Catholic God; Muslims have experiences of the Islamic god; Hindus have experiences of Hindu gods; Taoists have experiences of the eternal Tao; and so on. Obviously, these differing beliefs cannot all be correct – at least some of them must be wrong. But if some of them are wrong, why not all of them? What grounds do we have for supposing that one variety of religious experience is any more valid than any other? The answer is that we have none at all, other than an objective examination of the physical evidence to see if it backs up any religious group’s claims.
“Human existence is possible because the constants of physics and the parameters for the universe and for planet Earth lie within certain highly restricted ranges…. The ‘coincidental’ values of the constants of physics and the parameters of the universe point… to a designer who transcends the dimensions and limits of the physical universe.”
—Hugh Ross, “Design and the Anthropic Principle”, http://www.reasons.org/resources/apologetics/design.shtml
A variation of the traditional design argument, the “cosmological fine-tuning” argument holds that the fundamental constants of nature, such as the speed of light, the charge of the electron, or the value of the universal gravitational constant, are precisely tuned to permit life, and even a small change to any of them would have resulted in a universe drastically different from our own, one where no life could exist. Proponents of this argument claim this as proof that a benevolent creator set these values. Another version of this argument claims fine-tuning with regard to the solar system and the Earth, asserting that we could never exist if our planet had turned out slightly different in any one of a variety of ways.
The second form of this argument is easily disposed of. It is true that life as we know it, at least large, complex life such as human beings, would not exist on the Earth if conditions were significantly different (although the parameters of environmental variation are not as sensitive as some theists think; some, for example, assert that we would either freeze or boil if the Earth’s distance from the Sun varied by even a few kilometers, unaware that the planet’s orbit is an ellipse and its distance from the Sun varies by almost five million kilometers over the course of a year). However, this is not at all surprising. When we look into the cosmos, we see an enormous number of places utterly hostile to our kind of life: worlds bathed in the fierce glare of heat and radiation from nearby suns, worlds dark and frozen by the chill of space, worlds perpetually convulsed by volcanism, worlds constantly bombarded by impacting meteors, worlds lacking the chemical compounds necessary to sustain life such as ours. Is it any wonder that life arose here and not there? Is it any wonder that we live on the Earth and not one of those places? Of course not: we exist where we can exist. What would be far more miraculous, and far more indicative of supernatural influence, would be if we found ourselves on a planet where the environment was not conducive to our kind of life.
The cosmological version of the fine-tuning argument is more sophisticated, but nevertheless is not difficult to defeat. The key is to recognize that every step of this argument relies on implicit assumptions, none of which are supported by any evidence at all, and without any of which the argument collapses.
For example: how do advocates of this argument know how many different sets of values for the physical constants could have led to life? Perhaps life as we know it would not exist if these values were different, but that does not prove that life of any form could not exist. It could be that most of the universes that would result from changing these constants would contain some sort of complex life, even if that life would in most cases be extremely different from what we are used to. Advocates of the fine-tuning argument implicitly assume that only one, or at most a few, sets of values could have led to conscious life of any form, but there is no possible way they can know this.
Furthermore, how do advocates of this argument know that all possible values for these constants were equally likely, or even that they could have been different at all? Defenders of the fine-tuning argument implicitly assume that the current values were selected uniformly at random from an infinite or at least a very large range of possibilities, but there is no possible way they can know this. Indeed, for at least a limited subset of cases, we know that this is not true. Some physical constants are interrelated, such that a change to one would necessarily produce a change in the others. (For example, according to this Talk.Origins Post of the Month, the permittivity of free space, the permeability of free space, and the vacuum speed of light are related in such a way.) It is very likely that as physics advances, this process of unification will continue, and we will discover even more and deeper interrelations. A complete unified theory, the dream of physicists, might well show that things are this way because they could not possibly have been different.
Finally, how do advocates of this argument know that there was only one chance to get it right? On what grounds do they assume that the universe we can observe constitutes the totality of existence? Multiverse theories, which posit an enormous or even infinite number of parallel universes of which one is our own, are a common feature of theoretical astrophysics. Naturally, if there are thousands or millions of universes each with a slightly different set of values for the physical constants, it should be no surprise that we find ourselves in a universe where the values are conducive to our existence, rather than one where they are not. This is very similar to the rebuttal to the planetary fine-tuning argument discussed above.
As we have seen, there is no good reason to suspect the universe was fine-tuned. But consider this: even if there was fine-tuning, how do we know it was for humans’ sake? Human beings have a long history of considering themselves the centerpiece of creation, and an equally long history of being proven wrong on that point. On what grounds do we assume that we are the pinnacle of existence, or even that our existence was intentional? Perhaps the creator simply thought stars were beautiful and wanted a universe full of them, and hasn’t yet noticed that interesting things have begun to happen on some of the little bits of ash circling them. Or perhaps the creator did desire life to come into being but is not interested in human life. The evolutionary biologist J.B.S. Haldane, when asked what we could deduce about the personality of the creator by studying the creation, is said to have replied, “An inordinate fondness for beetles.”
“…belief in God need not be based on argument or evidence from other propositions at all…. [T]he believer is entirely within his intellectual rights in believing as he does even if he doesn’t know of any good theistic argument (deductive or inductive), even if he doesn’t believe that there is any such argument, and even if in fact no such argument exists…. [I]t is perfectly rational to accept belief in God without accepting it on the basis of any other beliefs or propositions at all. In a word… belief in God is properly basic.”
—Alvin Plantinga, “Is Belief in God Properly Basic?”, Nous, March 1981, p.41
A novel alternative to evidence-based arguments for the existence of God is philosopher Alvin Plantinga’s so-called reformed approach. Basing his theology on presuppositionalism, Plantinga draws an explicit contrast between what he calls classical foundationalism, which consists of the principle that belief should occur only where supported by the evidence and only to the degree that it is supported by the evidence, and his own approach, which is summed up by the above excerpt. Plantinga identifies certain propositions which are not believed on the basis of any other propositions, which he labels properly basic, and goes on to claim that God’s existence is one of these and therefore people are justified in believing it even if there is no evidence for it whatsoever.
The obvious rebuttal is to ask whether other propositions, such as the existence of Vishnu or Thor or the Great Pumpkin, or for that matter the nonexistence of God, could also be labeled properly basic and belief in them justified without any evidence. In response to this, Plantinga claims that not just any belief can be properly basic, but admits he has no criterion for deciding whether a particular belief should be properly basic or not. “…the Reformed epistemologist not only rejects those criteria for proper basicality, but seems in no hurry to produce what he takes to be a better substitute” (p.48). Instead, he suggests that this determination should be made inductively, by testing a belief against a set of examples agreed by everyone to be properly basic. When confronted with the obvious rejoinder that atheists and theists, or even Christians and other theists, will never agree about what examples are acceptable, Plantinga metaphorically throws up his hands and asserts that Christians are under no obligation to justify their beliefs to others. “Must my criteria, or those of the Christian communiy, conform to their examples? Surely not. The Christian community is responsible to its set of examples, not to theirs” (p.50).
Plantinga even goes so far as to assert that the existence of God is a properly basic belief because God exists, while the existence of the Great Pumpkin, say, is not a properly basic belief because there is no such being. “…the Reformed epistemologist may concur with Calvin in holding that God has implanted in us a natural tendency to see his hand in the world around us; the same cannot be said for the Great Pumpkin, there being no Great Pumpkin and no natural tendency to accept beliefs about the Great Pumpkin” (p.51). This is the height of circularity. Finally, he concludes that beings such as the Great Pumpkin are paradigmatic examples of “irrational basic belief” (p.51). But why should this make a difference? A properly basic belief, as Plantinga defines the term, is not based on any other belief at all, including beliefs about whether that belief is rational or whether one should believe rational things. If one postulates that a belief is properly basic, it should make no difference whether it is rational or not. For Plantinga to say otherwise is for him to abandon his own definition.
The “properly basic” criterion is hopelessly confused. It boils down to a naked assertion that there are some propositions we should believe without evidence, because Plantinga says so; and it is only these propositions we can believe without evidence, and not others, also because he says so. This does not in any way constitute an argument for the existence of God.
Are there such things as properly basic beliefs? Certainly an atheist could consistently hold this to be true. However, properly basic beliefs can never be about empirical truths; that type of knowledge always needs to be supported by evidence and is always provisional. Instead, I suggest, true properly basic beliefs fall into three categories: axioms that are true by the definitions of the system they reside in, direct mental experiences, and beliefs that are needed to make sense of the evidence we do encounter. Mathematical and logical axioms would be an example of the first category; we do not need evidence, for example, to prove that 1 + 1 = 2, or that the disjunction of a true statement and a false statement is itself a true statement. However, to the extent that these axioms are applied to objects and relations in the real world, rather than remaining mere abstractions, they should require evidence to support them.
The second category consists of beliefs about one’s own mental states, including emotions and direct sensory perceptions. While one can be mistaken about the cause of a particular mental state, or whether it corresponds to any actually existing object in the external world, one cannot be mistaken about the existence of the mental state itself. Plantinga himself suggests this in his paper. If I experience a pain in my right knee, I may be mistaken about why I am feeling it due to incomplete or faulty evidence, but I cannot be mistaken about the mere experience of feeling pain.
The third category includes propositions such as the effectiveness of induction as a means to understand the world, the scientific method as a reliable way to gain knowledge, and the existence of knowable laws of nature. None of these propositions can be proven by evidence, since they are higher-order propositions about evidence, but to discard them would leave us in a hopeless fog of uncertainty, since they are the only way to decide on the truth or falsehood of any other proposition independent of what anyone thinks. Therefore, we have little choice but to believe them, although happily they have proven extremely reliable so far.
However, the existence of God does not fall into any of these categories. God is not a being that exists by mere definition (as the ontological argument unsuccessfully tries to establish); nor is God a mere mental state, with no existence independent of people’s minds (at least, not according to most theists); nor is God’s existence a necessary proposition to make sense of the world (as the presuppositional argument tries and fails to claim). Therefore, we can safely conclude that God’s existence is not a properly basic belief, but rather an empirical proposition about the external world; it can only be properly supported by evidence, not mere assertion.
See “A Flip of the Coin“.