The Quantum Thomist

Musings about quantum physics, classical philosophy, and the connection between the two.
Intellectual conservatism interview.

Is God a failed Hypothesis? Part 4: Philosophical Objections
Last modified on Thu Sep 3 20:16:56 2020


This is the fourth post in a series on Victor Stenger's book God: The failed hypothesis. In this post, I intend to partially review his first chapter. I discussed most of the chapter in the previous post, but the chapter also contains a few "logical disproofs" of God, summaries of various arguments from other writers. My intention in this post is to briefly discuss those.

Professor Stenger himself is not very confident that these arguments will convince theists. His reason for saying this is that believers can always change their definitions of God in order to find and exploit a loophole in the argument. My own feeling (as I shall argue) is that he has this the wrong way round; that the framers of the arguments misunderstood what theists have believed about God, in order to construct an argument against Him. Most of the arguments presented attack the sort of God advocated for by Kant. Often they then spoil this by not working in a Kantian framework, bringing in additional assumptions which Kant would not have accepted. But I don't advocate for Kant's model of God or His philosophy, instead preferring the classical understanding of God of orthodox Christianity. Certainly there are minor differences in the conception of God between Eastern Orthodoxy, Roman Catholicism and classical reformed Protestantism. But all of them stand in the tradition of Platonic-Aristotelian philosophy, interpreted to bring it into line with Hebrew and early Christian monotheism. The differences between these strands of Christian theism are minuscule in comparison to the gulf between them and the post-enlightenment visions of God which Professor Stenger's arguments address.

Of course, the arguments don't originate with Professor Stenger. He drew them from an anthology edited by Michael Martin and Rikki Monnier called The impossibility of God. This anthology collects together what it regards as the most important arguments from atheist philosophers of religion purportedly showing that the usual definition of God is contradictory. These aren't New Atheist hacks like Victor Stenger, Lawrence Krauss, and Richard Dawkins, whose ignorance is manifest in every word they write. The essays within the collection are from among the best and most knowledgeable minds in the atheist ranks. These are people to be respected and reckoned with.

I recommend this book to all intelligent and educated classical theists. It will strengthen your faith no end. There is little more satisfying than coming up against the best arguments the opposition has to offer, and one by one defeating them.

Professor Stenger only discusses a selection of the arguments in his book, presumably either those he felt the strongest, or those which he thought could be most clearly appreciated by his target readership. I'll restrict my discussion to just those arguments which Professor Stenger references, referring back to the original essay as appropriate.

I'll defer my discussion of one of the arguments which Professor Stenger lists, namely the problem of evil. This is because Professor Stenger discusses that argument in more detail in a later chapter, and I would rather discuss it in depth when I come to that point in the book.

An all-virtuous being cannot exist

This argument is based on the essay by Douglas Walton, who expands an argument by the Platonic philosopher Carneades, but expressed in more modern terms.

  1. God is (by definition) a being than which no greater being can be thought.
  2. Greatness includes greatness of virtue.
  3. Therefore, God is a being than which no being can be more virtuous.
  4. But virtue involves overcoming pains and danger.
  5. Indeed, a being can only be properly said to be virtuous if it can suffer pain and be destroyed.
  6. A God that can suffer pain or is destructible is not one than which no greater being can be thought.
  7. For you can think of a greater being, that is, one that is non-suffering and indestructible.
  8. Therefore God does not exist.

The first objection I have to this argument is the definition used for God. Walton draws this from Anselm's ontological argument; and this definition was picked up on by Kant and his successors. It is not, of course, the definition which a classical theist will use, which instead arises from the cosmological argument. The consequences of the classical theist argument is that God is immutable (which implies indestructibility) and lacks passions (which implies non-suffering). These are built into the classical conception of God; they can't be avoided. So a suffering or destructible God is viewed as a contradiction.

But, of course, classical theists also tend to view God in terms of perfection, and one can see how this might lead to the premise of God being the greatest possible being. God's perfection is directly implied from His definition in terms of the Cosmological argument, and from His lack of potentiality. There is, of course, a difference between the greatest possible being and the greatest conceivable being. Firstly, there are many conceivable beings which are impossible. I can imagine a magic killer rabbit of Caerbannog capable of decapitating knights with a single leap, but that doesn't mean that such a being can logically exist. On the other hand, it might equally be true that there are possible beings which are not possible for us to grasp fully in our imaginations. The God of classical theism is the standard example.

However, the biggest problem I have in this definition is in the concept of greatness. What does it mean for something to be great? For the argument to work, greatness has to include the concept of virtuous, or (more broadly) goodness. And here (as a classical theist) I encounter the problem. What it is for something to be good depends on the nature of that thing. A perfect apple is not the same as a perfect fireman. In each case, the qualities implied by perfection are different between the two different beings. For example, perfection in an apple might include the qualities of being crunchy, sweet and nutritious. None of those are implied when we think of the perfect fireman, nor would any of these attributes assist him as he tries to put out fires.

In the same way, for God to be perfect means something different from what it would mean for a man to become the perfect example of humanity. We discuss the virtues as individual elements of a good character in mankind, but divine simplicity makes it harder to separate the attributes of God. So if the word "virtue" can usefully be applied to God, it certainly isn't in the same sense that the word is applied to man. Furthermore, when we apply the word used to describe a human virtue such as knowledge or power to God, we don't use the word in the same sense. God's knowledge isn't seen as the same sort of thing as human knowledge, just differing in the amount of knowledge God has, but a wholly different sort of thing. There is a discontinuity between the two concepts. The word knowledge is used analogously rather than directly (or univocally).

Equally, there are many virtues in man which simply don't apply to God. Temperance is one example; the habit of taking food and drink (for example) in a healthy quantity. Having the inclination to neither overeat nor starve oneself. Clearly God has no need for food nor drink, so this human virtue is not part of God's perfection. There is no problem with that, since God is different from humanity, so the sort of attributes which make God perfect will be different from those which make a man perfect. There will be some aspects of divine perfection which have no analogue in mankind, and some aspects of human perfection which have no relevance to divine perfection.

Clearly, then, it is the third, fourth and fifth premises of this argument which are most troubling. The third premise because it discusses virtue in general and not in terms of the being that the virtue is applied to. This will lead open to a fallacy of equivocation. The fourth and fifth premises make claims which might be true for human virtue, but then applies them to divine perfection. I do have issues with the fourth premise even as applied to humanity. Virtue is about one's character, not one's actions. Being courageous means that if one encounters pain or danger, then one would (if it were not reckless to do so) be able to overcome it. But that "if" need not be satisfied, and it makes no difference to whether someone is courageous. To exhibit bravery, one needs to a) to have the virtue of courage; and b) to be placed in a circumstance of danger. These two conditions are independent of each other, one referring to the inner character and the other to the external situation. One can still have the virtue of courage even if one is perfectly safe at home. One can still have it even if one has never faced danger. Obviously, nobody will know if you are courageous until you face danger, but this is different from the question of whether or not you are truly courageous.

But the obvious response to this argument is that whatever its merits with regards to human courage, courage is, like temperance, simply not applicable to the divine nature. It is a contradiction to call God either courageous or cowardly. And because the list of divine virtues (if we can use that term) depends on the divine nature, saying that courage is not among them takes nothing away from God.

So how does Walton defend this? In his defence of the argument, he focuses on premises four and five, focusing on the virtue of courage. He lists the cardinal virtues of courage, temperance, justice and prudence, and uses definitions from Cicero. He notes that "All four cardinal virtues are distinctively human in that all of them require balance in an uncertain and difficult situation, where hardships and dangers -- either physical dangers, or the danger of making a bad decision under pressure -- need to be overcome or at least dealt with."

He proposes a few ways out. Firstly, to deny that courage is a cardinal virtue. Secondly, to redefine the meaning of courage (although all his proposed definitions refer to actions in the face of danger rather than the state of mind that is implied by the virtue). He is aware of Aquinas's statement that virtue and wisdom are not predicated of us and God univocally. But he doesn't make anything of this, except to say that Western theology has been well aware of the sort of objection addressed here, and has incorporated the answer to it into its description of God.

This statement by Aquinas invokes ideas that were part of my objection to the argument, but Walton didn't answer the problem it poses for his argument. Nor does he answer the objection that he is equivocating over the meaning of the word "virtue". He defends (by quoting authority, so not especially well) that the cardinal virtues are indeed aspects of human goodness, and I agree with him. But his argument relies on the proposition that these virtues are aspects of what it means for God to be perfectly good. He does not defend this proposition. Thus his argument fails.

Of course, the classical construction of God contains no contradiction. Perfection in God means perfection in everything which is logically consistent with the divine nature. If courage, recklessness and cowardice are inconsistent with the divine nature, then they are not aspects of divine perfection. But, then, is there any theist who ever claimed that they were?

I should make mention of the incarnation here, which might also be relevant to the argument. As the second person of the Trinity took on flesh, God did face danger and hunger. However, I'm not sure that is the sense in which the objection was made, so I won't dwell any longer on this point.

God as a fitting model of worship.

This argument is taken from an essay by James Rachels.

  1. If any being is God, he must be a fitting object of worship.
  2. No being could possible be a fitting object of worship, since worship requires the abandonment of one's role as an autonomous moral agent.
  3. Therefore, there cannot be any being who is God.

This argument is expanded greatly by Rachels in his essay. The obvious point of contention is the second premise, but there might also be issues in the first premise, since Rachels might not define worship in the same way that a Christian would. How would a Christian define worship? It's not a simple concept, and contains various aspects to it. Augustine discusses worship in detail in his first chapers of his City of God. For example, in City of God, 10.1, Augustine relates worship to honour of God, piety (which includes works of charity), service, and religious ritual.

The first part of Rachel's essay relates to the definition of worship. He dismisses the definition that worship is to bow down with silent awe. Worship involves the acceptance of various beliefs about God. These beliefs are not accepted with indifference. And they influence his self-image and way of thinking about the world. So, for example, the believer will seek out the will of God, and to consult authorities, pray, and reads scripture. The believer will accept himself as a child of God, and of his conduct as offering either honour or dishonour upon God.

Rachel also relates worship to something which is done. These actions are ritualistic in nature. A ritual contains a ceremonial aspect, and also something which it is meant to accomplish. The various aspects of the ritual serve to symbolise something about the relationship between the worshipper and the worshipped. In particular, in worship, the worshipper is acknowledging and accepting his role as a child of God. It assumes the superior status of the one worshipped. The worshipper assumes his own inferiority, which involves an attitude of humility.

The function of worship of glorifying or praising God is derivative of this role of accepting and acknowledging inferiority.

Roles can be violated. To violate the roles of being a worshipper would include disobeying God, or setting ourselves against God.

Thus to say that a being is worthy of worship, we recognise Him as having an unqualified claim on our obedience. Could there be such a claim? There is a long tradition in moral philosophy, from Plato to Kant, that that recognition can never be made by a moral agent. In this tradition, a moral agent needs to be autonomous, or self-directed. Moral precepts are imposed by the agent upon himself. "The virtuous person is therefore identified with the person of integrity, the person who acts according to precepts that she can, on reflection, conscientiously approve in her own heart." On this view, to deliver oneself over to a moral authority is incompatible with being a moral agent. It makes no difference what that authority is.

There is then a conflict between the role of worshipper, which commits one to total subservience to God, and that of the moral agent, which involves autonomous decision making.

That's about as far as Rachels gets, which is troubling to me since I don't yet see any contradiction. One could say that the worshipper of God is no longer a autonomous moral agent (or has surrendered his moral autonomy), and be perfectly happy with that circumstance, regardless of what Kant and others wrote. So I will try to extend the argument to find a possible contradiction.

The argument, as I read it, is this. If God is to be a fitting object of worship, then it must be morally right to worship God. But moral rightness and wrongness are only valid concepts for autonomous moral agents. Worship denies moral autonomy, and thus that there can be any ethical decision to make. Thus it cannot be morally right to worship, since the act of worship denies morality. This contradiction can only be resolved by denying that anything is a fitting object of worship.

Before proceeding, I ought to quickly outline the argument that if God exists, then he is a fitting object of worship (as Rachels defines worship). If God exists, then it is true that we are inferior to Him. It is good for us to accept and live according to the truth. Therefore, if God exists, then it is morally right for us to acknowledge and live according to the fact that we are inferior to Him. Inasmuch as this is part of worship, then it is morally right for us to worship God. And if it is morally right for us to worship God, then God is a fitting object of worship. Clearly this argument needs to be fleshed out more, (there are a couple of places where I jump in my terminology) but that is the outline of it. The place where I have put in a moral judgement is in the premise that it is good for us to accept the truth. But this strikes me as self-evident. If we suppose that the contrary is true, then it would be true that it would not be good for us to accept the truth. But that includes the truth that it is not good for us to accept the truth. Thus we find that it is not good for us to accept that it is not good for us to accept the truth. This is a form of the liars paradox. Thus saying that it is not good to accept the truth is incoherent, from which we conclude that it is good to accept the truth. [Of course, one could try to get around this by saying that we should accept some truths and not others, but then there comes the problem of distinguishing between which truths to accept, which would ultimately lead to a similar problem.]

So what do we make of Rachels' argument? His understanding of worship isn't quite how I would express it, but there is nothing in his discussion where I find myself disagreeing. His discussion might be incomplete; for example there is a sense in which acts of piety are part of worship, that is to say the decision to choose to pursue goodness towards our neighbour as well as service and humility before God. One criticism that I would make is that his discussion only related to Christian worship. For example, the emphasis on becoming children of God is a specifically Christian concept. But one can worship gods other than that of Christianity.

I would also question whether worship necessarily implies accepting inferiority. For example, in the Church of England marriage service, the groom vows to the bride,

With this ring I thee wed. With my body I thee worship. And with all my wordly goods I thee endow.

In this case, worship is not meant in the sense of declaring one's inferiority, but is instead a promise to show honour, respect and adoration to the wife. But, obviously, worship of God is somewhat different, and I won't suggest that the Christian worshipper will deny God's superiority.

The obvious response to Rachels' argument is that it depends on an understanding of ethics which is wholly subjective. Our actions are good if and only if, in retrospect, we ourselves approve of them. But this just leads to the age old problem of moral relativism. If goodness is defined in terms of conformity to our own desires, the how do we judge whether or not those desires are good? Most theists will instead regard moral standards as objective. This is regardless of whether they accept Natural Law ethics, divine command, or a combination of them both. The question is not just whether we approve of the action, but whether or not the things we morally approve of are in line with that external objective standard. If the external moral standard states that it is good to submit oneself to God, then we ought to bend our own moral values so we regard submission to God as a moral good.

And this is true of any objective moral standard. As soon as we accept some set of moral values, whatever they are, we loose our moral freedom. Unless we later change our mind -- but the worshipper of God is also free to later change his mind.

To be good means conforming our desires and attitudes to that standard. If this is true, then submission to God is not the denial of moral autonomy, but the fulfilment of the moral life. For part of morality is to accept the truth. Denying our inferiority to God is thus an evil. To fail to worship, in Rachels' sense, is a bad moral choice. That objective standard just needs to exist to make God a fitting object of worship. It doesn't matter whether or not we agree with it. It doesn't matter if nobody agrees with it, or fully understands it. If there is an objective standard, and if that standard states that it it is morally right to worship God, then it is right to do so regardless of whether there are any autonomous agents or what they think, or whether or not they choose to surrender their autonomy by worshipping.

Worshipping God is not just a simple abdication of moral choice. At each moment, we have to make a decision. Ultimately this decision is whether to honour God or dishonour him. But practically, the scope of each decision is smaller. Do we pick up that litter, or leave it in the street? Do we give to the beggar, or walk past him? Rachels' argument seems to suggest that once we decide to follow God we lose all moral autonomy. But that is not the case. Following God merely means resetting our moral compass; aligning our priorities in line with God's. But we still face the same decisions and choices as Rachels' morally autonomous person. The only difference is that there will be cases where the follower of God will declare one attitude to be good, while the morally autonomous person will declare the opposite action as good. The only question of importance is which of these two people is correct? The one which conforms to the objective standard of morality.

Thus Rachels' argument fails because it presumes a secular notion of ethics which no classical theist would accept. The contradiction is not between the idea of ethics and the idea of worshipping God, but because Rachels is bringing in an ethical standard from one philosophical tradition and trying to balance it against an understanding of God from a contradictory philosophical tradition. The contradiction is in his merger of different world-views, not in the notion of God.

Rachels lists various possible objections to his argument.

  1. What if God issues no commands other than to obey our consciences? This objection is irrelevant, since a) that's not what theists believe; and b) it depends on the same primary error of the argument, namely the assumption of moral relativism.
  2. God is perfectly good. In obeying God, we would only be doing what we ought to do in any case. Rachels' response to this is that being perfectly good is an essential aspect of being God. Therefore, we cannot judge a being to be God unless we first judge Him to be perfectly good. But this requires knowing that his commandments are right. There is circularity here, which Rachels objects to. Obviously, there are two ways out of Rachels' circle. The first is to identify God by some other means than His perfect goodness, such as from His being a first cause, or the one who raised Jesus from the dead. We then need not make any moral judgement before recognising God as God. Secondly, we can reason our way to an objective understanding of moral values semi-independently of God. For example, by using secondary causation and natural tendencies to define purpose and thus goodness. I say semi-independently, because accepting these premises leads to the acknowledgement of the existence of God through Aquinas' fifth way, and ultimately the secondary causes and tendencies arise from God's plan and design; even if we can come to knowledge of them through observation and reason. The classical theist would accept both of these approaches. Again, Rachels' response to this objection seems to be resting on the assumption that moral values are subjective, and that we have no independent way of discovering an objective moral standard.
  3. People are sinful. Their consciences are corrupt and unreliable guides. Not being able to trust our own judgement, we have to trust God and do what he wills. Rachels' response to this objection is that it is inconsistent. If our judgement is corrupt, then how can we know that our judgement is corrupt? Even if we trust God to tell us what is right, how do we know that that judgement to trust God is correct?

    This objection again problematically assumes that our moral standards arise from our subjective judgement; so it is not an objection which I would personally make. Even so, there are several instructive flaws I can see with Rachels' thinking as he delivers his response. The first is that sin is not just a matter of our judgement of what is right and wrong. It partially involves breaking the moral standard while remaining aware and conscious of it. Perhaps this isn't so relevant, since the objection refers to our imperfect judgement of what the moral standard is. But that we can't keep to even our own standard raises our consciousness that we are flawed. If we ever change our mind on a moral issue, that too demonstrates our fallibility. Thus we can have good reasons for suspecting that our judgement might be flawed, even though we might not be able to pinpoint exactly where. Secondly, that our judgement is flawed in some respects does not mean that it is totally corrupt. We may know enough to be able to work out at least some of the objective standard through reason; enough to point us to our inadequacy to compute the rest of the moral law and thus our need for God. Thirdly, we can consult other authorities about the moral standard. Maybe we have to judge which authorities to trust, but this isn't the case of just arbitrarily choosing. An authority who logically argues from unavoidable premises to firm conclusions is obviously more reliable than one who plucks his axioms out of thin air, or who advocates obviously incoherent positions such as moral relativism or the belief that it is objectively true there is no objective truth. Fourthly, different people come to different, and perhaps contradictory, moral judgements. That some people's judgements are flawed does not mean that everyone's is flawed. Just as although some people compute an incorrect answer to an algebraic problem does not mean that everyone gets the answer wrong; nor that those who get it right cannot be confident that they have done so. And of course, there are many moral questions. Somebody might be correct in one area, but wrong in another. If we find our moral judgements disagree with another person, then that implies that at least one of us is wrong. Finally, just because our moral judgement might be flawed, that doesn't mean that our judgement in other areas is flawed. For example, I might not be confident about my judgement in a difficult moral quandary, but perfectly happy with it in matters of arithmetic or in evaluating historical evidence or determining that something is miraculous. In particular, if there are independent reasons for trusting God, apart from moral judgement, then that trust will naturally then be extended to cover ethical issues.

  4. Some have argued that the voice of the individual conscience is the voice of God speaking to the individual. I think a classical theist might have a little sympathy for this view (God can prompt us through a conscience), but few would say that all human conscience is always the result of God's prompting. Thus I would not use this objection to Rachels' main argument.

    Rachels' response to this objection is based on the old Euthyphro argument. If God's prompting of our conscience is right, then there is no need to think that we couldn't discover this for ourselves. In that case, God would be redundant. If, however, it is arbitrary, then the notion of the conscience is a sham. This response has the flaw that it fails to recognise that if we can uncover the true moral law for ourselves -- if the moral law is built on objective premises that we can apply reason to -- then we still have to discover those principles, and ultimately they would depend on human nature and thus God's choices in his design of creation. We might need God's promptings to find the right moral premises -- maybe some of them aren't things we can discover on our own. And even for those we can uncover from reason applied to observed facts, God is still not redundant because God's initial choices in creation are the ultimate cause of those observed facts. God would select the moral standard as part of his design for mankind.

  5. Is the question of whether a being is worthy of worship different from that of whether we should worship him? I would say no, and agree with Rachels response to this objection.

So Rachels never addresses the main objection to his argument -- namely that it depends on a definition of moral autonomy derived from Kant, which in turn depends on the idea that ethical standards are merely a matter of our own subjective judgement. Every classical theist will believe in some form of objective moral standard, against which our own values are judged. If we are to find a contradiction in the definition of God, we need to use an understanding of ethics that is consistent with that notion of God. In other words, the ethical principles accepted by a classical theist, not those accepted by Kant.

Our moral value as autonomous moral agents is in whether or not we choose to align our desires to that objective moral standard, and how well our actions are in accordance with those moral values. To be good, the autonomous moral agent needs to choose to align his values to that external standard. Everything else falls short; everything else is an evil. If by declaring ourselves as autonomous moral agents, Rachels means that we have to make moral choices, then worshipping God in no way undermines the need to constantly make those choices. There would be no contradiction between worship of God and our status as autonomous moral agents in this sense. If, on the other hand, Rachels understands an autonomous moral agent to be someone who defines their own moral standard, then our ability or lack of it to do that is irrelevant to the question of whether it is objectively morally right to worship God. God's status as a fitting object of worship depends on the objective moral standard, not any fleeting judgements we might make. Whether we are autonomous moral agents or moral robots makes no difference to the external moral standard that declares God as a fitting object of worship. Still less the question of whether we ought to be autonomous moral agents, or fixed in our moral values. Thus whether we are morally autonomous in the sense that Rachels argues for (i.e. we can decide for ourselves our moral standards) is irrelevant to the question of whether or not worshipping God is objectively morally right. Morality does not depend on our subjective judgements of moral standards; but instead those judgements are judged against the divine standard, and in Rachels' case, found wanting.

The perfect creator

This argument is taken from an essay by Theodore Drange. Drange, unfortunately, does not give more than an outline of the argument, so it is difficult to properly engage with it. But here is the outline:
  1. If God exists, then He is perfect.
  2. If God exists, then He is the creator of the universe.
  3. If a perfect being exists, then whatever he creates must be perfect.
  4. But the universe is not perfect.
  5. Therefore, it is impossible for a perfect being to be the creator of the universe.
  6. Hence, it is impossible for God to exist.

Drange's only comment on this is to make mention of the response that the universe was made perfect, but subsequently spoiled by the acts of man. He mentions that this raises issues that arise in the deductive argument from evil, and does not discuss it further.

So that is all we have to go on in discussing this argument. As a Christian, I have some sympathy to the response that Drange mentioned. The idea of our universe being created very good and subsequently marred and in need of healing is an essential part of Christianity. However, I will accept that this argument by itself is not a complete response to Drange's challenge. After all, whatever marred the universe is a part of creation, so it was presumably originally created perfect, or was created by something originally created perfect. And if so, by Drange's premise three, it would not produce any work which is imperfect. Thus, I think if we accept the terms of Drange's argument, this response would not by itself be enough to defeat it.

I think the two places where this argument falls short are in its notion of perfection and in its denial of the possibility that a perfect being could produce an imperfect universe.

We have to ask what we mean when we say that something is perfect, and I think that the meaning is always understood in terms of some good or end. Something is perfect if it completely fulfils that good. Here we run into the first problem. Just as the word goodness is not univocally applied to both God and Creation, but only analogously, so to is the concept of perfection. Thus when we discuss the perfection of God and the perfection of a creature, we mean two different things. In particular, the word "perfect" is written down in two different senses in premise 3.

One way of thinking about goodness is in terms of states of being. A changeable being can exist in numerous different states. One of those states is actual at any given time (if we neglect superpositions for simplicity, although this model can be enhanced to allow for that). Now to be good with respect to some particular end entails being in a particular subset of those states, those which are able to fulfil that end. That is one of those subset of states is actual. If we think of different goods (or different natural ends or tendencies), then each of those would imply a different subset of states. Perfection would be the intersection between those subsets. That at least, is how we might understand perfection in a creature. In particular, to force a being into its perfect state is to severely curtail its capacity to change.

Aquinas defined divine perfection in terms of God's pure actuality:

Hence, the first active principle must needs be most actual, and therefore most perfect; for a thing is perfect in proportion to its state of actuality, because we call that perfect which lacks nothing of the mode of its perfection.

So divine perfection is not defined in terms of an end, but arises because God is incapable of change, and thus incapable of being imperfect. That is not true for any created being.

In particular, we can think of a being created with the tendency to grow towards maturity (whether physical or intellectual). Growth entails change. Since this is growth towards some end, it follows that not all stages of the beings growth fully exhibit that end. Thus any such created universe must at some point in time be distant from a state of perfection. The state of perfection might some day be reached, but it cannot be there at the start. If God created the universe with everything mature, then the beings would not have been able to exhibit their end of growth. In this case, the idea of a universe that is perfect throughout its existence is incoherent. God could choose a universe which will grow into perfection (whether naturally, or, as in the case in our universe, as a consequence of divine grace), but such a universe would at times be imperfect. Equally, during the period of imperfection, there might be struggles between different beings each seeking their own perfection, leading to various dead ends and occasions where the capacity for growth is not realised. Growth towards perfection need not be a linear or straight-line process.

All we need to complete the response is to allow for the possibility that God ranks an universe where beings have the capacity for change and growth as superior to one in which they don't. And I think it is impossible to argue that this is definitely not the case. We can identify goods through reason and consideration of natural tendencies or final causes, but it is much harder for us to rank those goods in order of priority. We can't say that a universe with one set of ends is better or worse in God's eyes than a universe with a different set of ends.

A similar argument is often made in terms of free will. I need not outline the details, because they mirror those of the argument above.

So I don't think that premise 3) of the argument is valid. The only being which is perfect essentially is God. Any created being must carry within it the possibility of imperfection, that is it must have potentiality. So it is not logically coherent to say that God could create a perfect universe, using "perfect" in the same sense that is applied to God. And if we think of "perfect" in the other sense, that is exhibiting a state of goodness in a universe that has the potential to be not good, a universe fixed in that state need not necessarily be the best possible universe, in the sense of the universe that displays the most goodness. It might be that an imperfect universe which allows for change is seen as better than the perfect static universe.

Transcendence and omnipresence

This argument also comes from Drange's essay. Again, it has limited exposition, so we largely only have the outline.

  1. If God exists, then He is transcendent (i.e. outside space and time).
  2. If God exists, then He is omnipresent.
  3. To be transcendent, a being cannot exist anywhere in space.
  4. To be omnipresent, a being must exist everywhere in space.
  5. Hence, it is impossible for a transcendent being to be omnipresent.
  6. Therefore it is impossible for God to exist.

Drange considers objections to premise 3, saying that something could transcend space by being partly inside space and partly outside it. However, it is premise 4 where I would question this. Once again, this is related to the distinction between God and creation. We think of an extended object as occupying a certain region of space. The larger the object, the more space it occupies. Then extrapolate to infinity, and you might think that you get omnipresence. That is what it would mean for a creature to be omnipresent, and I think that is the sense in which Drange understands the term. But this obviously this isn't how a theist will think of God's presence. The word is again used analogously rather than univocally.

So, how would a theist understand omnipresence? In fact, Aquinas anticipates this same objection.

Objection 1. It seems that God is not everywhere. For to be everywhere means to be in every place. But to be in every place does not belong to God, to Whom it does not belong to be in place at all; for "incorporeal things," as Boethius says, "are not in a place." Therefore God is not everywhere.

So, we can see Aquinas' answer

I answer that, Since place is a thing, to be in place can be understood in a twofold sense; either by way of other things i.e. as one thing is said to be in another no matter how; and thus the accidents of a place are in place; or by a way proper to place; and thus things placed are in a place. Now in both these senses, in some way God is in every place; and this is to be everywhere.

First, as He is in all things giving them being, power and operation; so He is in every place as giving it existence and locative power. Again, things placed are in place, inasmuch as they fill place; and God fills every place; not, indeed, like a body, for a body is said to fill place inasmuch as it excludes the co-presence of another body; whereas by God being in a place, others are not thereby excluded from it; indeed, by the very fact that He gives being to the things that fill every place, He Himself fills every place.

And his answer to the objection,

Reply to Objection 1. Incorporeal things are in place not by contact of dimensive quantity, as bodies are but by contact of power.

In other words, there is a difference in what we mean when we say that an incorporeal being is in a place and a corporal being is in a place. There are perhaps three senses to the word discussed by Aquinas here. Firstly, there is the sense in which a corporal has dimensional extension over time and place (contact of dimensive quantity). Then, we have the ability to interact with other beings (contact of power). Then we have the idea of giving being to something in a particular place. The first of these senses is the one which Drange considered. He neglects the other two. A corporal being can be in a place in the first two of these senses. The last two apply to God. Thus when we talk about God being omnipresent, theists do not mean that God exists in every place, but can interact with and give being to things wherever they happen to be. And in this sense, omnipresence is clearly not inconsistent with transcendence. Indeed, only something transcendent could be omnipresent in this sense, since nothing in space and time is infinite.

Thus Drange's argument fails in its premise 4, because he misunderstands what classical theists mean when they discuss God's omnipresence.

Personality and Physicality

This is another argument taken from Drange's essay, and like the others he provides few details. The difference in this case is that even Drange doesn't seem to be convinced by it.

  1. If God exists, then He is nonphysical.
  2. If God exists, then He is a person (or a personal being).
  3. A person (or personal being) needs to be physical.
  4. Hence, it is impossible for God to exist.

Clearly, premise 3 is the important one here, and looks to be an example of begging the question. In justification, Drange only gives a reference to a work by Kai Nielsen which I don't have access to. I do, however, have access to and have read some of Kai Nielsen's essays, and this one here seems to be on the same topic. So I will discuss the argument as presented in that paper. If the version of the argument in his Introduction to the philosophy of religion differs from this, then please tell me, and I will endeavour to enhance this response to address that argument as well.

The first argument presented by Nielsen is this one:

The Christian goes on to say that this individual is immaterial, then incoherence, so Daher avers, results. The reason is quite simply that since 'individual', as it is ordinarily used, implies 'identifiable', and 'immaterial', as it is ordinarily used, implies 'unidentifiable', the belief that something E is both an individual and immaterial would entail two incompatible propositions, namely 'E is identifiable' and 'E is not identifiable'. Such an argument, recalling that a non-anthropomorphic God is allegedly immaterial and non-identifiable, shows that the concept of God is incoherent.

The obvious response to this is that, to the classical theist, being immaterial in no way implies that it is not identifiable. True, material beings are distinguished in part by their matter, and in part by their form. But immaterial beings can still be distinguished from each other (and thus identified) through differences in form. This objection thus seems to be resting on a philosophy which denies that form can be used as an identifying principle.

Nielsen responds to this objection with the following arguments:

  1. In identifying organisms other than persons bodily criteria are used. It is certainly true that in identifying some creatures other than persons bodily criteria are used. But in this no way proves that it is true for all creatures, whether persons or not. Since this is the very point under discussion, this response begs the question.
  2. In speaking of God we are talking about a sui-generis reality of which, by definition, there could not be more than one such reality. But, given the breakdown of the ontological argument, and with it the breakdown of the claim that 'There is a God' is analytic, there can be no purely 'conceptual identification' of God. This assumes that any purely conception identification of God falls back on the ontological argument. I suspect this is in reference to Kant's discussion. But Kant's argument that (for example) the cosmological argument collapses into the ontological argument depends on his definition of God that arises from the ontological argument. But why should we accept this definition? Aquinas, for example, rejected the ontological argument, but still had a clear identification of God, defined in terms of his versions of the cosmological argument and teleological argument. Whatever non-ontological argument, or revelation, you use to understand or identify God, any failure of the ontological argument will be irrelevant to it. Nielsen asserts that we have no intelligible directions to identify a God beyond any purported created gods of polytheism. But this is just an assertion, which theists will disagree with.
  3. If we have a principle which covers a myriad of different instances and gives a powerful explanatory account of why they are as they are, we would, when faced with an anomalous instance or even a cluster of such instances, try (as Stephen Toulmin puts it) 'to save appearances' and account for the anomaly without abandoning the principle. ... It is far more plausible to hold to the principle which would justify classifying them as incoherent or devoid of factual significance or content, rather than to abandon the principle. This is simply bad practice. When a scientist encounters new evidence or new facts which stand against an established theory, he either modifies the theory or replaces it with a better one. There is no need to abandon the principle that we distinguish between different individuals by matter; we merely need to extend it to say that we can distinguish between different individuals either by matter or by form (or perhaps both).

In summary, I don't see that Nielsen offers any good support for his contention that non-material beings are not identifiable. As with many of the examples listed before, he fails to properly appreciate the creator/creature distinction.

Aquinas discusses the topic in question 29 of the Summa Theologica, and in particular article 3 addresses a similar objection. His answer is that there are different senses in which we can describe something as an individual. We cannot say that God is an individual through distinct matter, but individuality also means incommunicability, and in this sense we can distinguish God from other beings.

All this has to be said with the priviso that I haven't access to the original source for the claim, and that there might be a stronger defence of premise 3 elsewhere that I am unaware of. The arguments in favour of premise 3 I am aware of all fail, and thus I believe that this argument should be rejected.


Stenger's final argument is taken from an essay (or rather two essays) by J.L. Cowan.

  1. Either God can create a stone which He cannot lift, or He cannot create a stone which He cannot lift.
  2. If God can create a stone which he cannot lift, then He is not omnipotent (since he cannot lift the stone in question).
  3. If God cannot create a stone which He cannot lift, then He is not omnipotent (since He cannot create the stone in question).
  4. Therefore, God is not omnipotent.

First of all, we need to define omnipotence. Cowan begins his first essay with a definition,

The claim that God is omnipotent is presumably the claim that He can do anything.

However, he qualifies this with the qualification that the "anything" in his statement excludes things which are obviously self-contradictory, such as squaring a circle.

Aquinas' definition at first sight seems to be similar,

All confess that God is omnipotent; but it seems difficult to explain in what His omnipotence precisely consists: for there may be doubt as to the precise meaning of the word 'all' when we say that God can do all things. If, however, we consider the matter aright, since power is said in reference to possible things, this phrase, "God can do all things," is rightly understood to mean that God can do all things that are possible; and for this reason He is said to be omnipotent.

So far, so much in agreement with Cowan.

But there are a few additional qualifications. Firstly, there is what is meant by possibility. This can either mean possible for some being (or a relational possibility), or possible in some absolute sense. God cannot be said to be able to do all things that are possible for created nature, because God's power extends further than that. In this way, extrapolating from the power of created things to God's power is a mistake. Nor can we say that God is capable of doing all things that are possible for God, because this would just be circular. Thus, the statement must be that God can do all things possible absolutely, i.e. those things which don't contain a contradiction, or which are incompatible with the predicates.

The second distinction that Aquinas makes is between active and passive power. Active power is the principle of being able to act upon something else, while passive power is the principle of being acted upon. God only has active power, since passive power implies some possibility of change, which God lacks. So God's omnipotence only refers to actions on other types of being, and not to, for example, the ability to move Himself or sit down.

The obvious answer to Cowan's argument is to say that the notion of a stone that is impossible to lift is incoherent. We can of course object to this particular example though an examination of the basic physical principles. When we discuss lifting, we mean raising something up against the force of gravity. We also imply a certain degree of proximity between the object we are lifting and the source of the gravitational field. We don't discuss lifting something of the earth if that object is in a distant corner of the universe. The amount of energy required to lift something depends on the mass of the object, the strength of the gravitational field (i.e. the mass of the object generating the gravitational field), and the height which we lift it by. For something to be impossible to lift to any height would mean that an infinite amount of energy is required to lift it, which either implies that the object itself, or the object generating the gravitational field, has an infinite amount of mass. This is, however, absurd; firstly because nature abhors infinity; secondly because it would collapse into a black hole, and thirdly because that black hole would suck everything in proximity with itself into itself, devouring everything in its neighbourhood. In such a universe where there was such an infinite mass black hole, that black hole would soon become the only being in the universe. But the notion of lifting requires two beings: the one generating the gravitational field, and the one that we are lifting. Thus the notion of a stone that is impossible to lift is incoherent within the universe in which we live. The argument fails in the same was as if God were required to create a square circle.

Could we avoid this objection by imagining a universe with a very different physics to our own? But such a counter-argument would only be effective if such a universe were self-consistent and the sort of universe that it is logically possible for God to create. If the argument from omnipotence succeeds in that universe, then it is not logically possible for God to create that hypothetical universe, and in that case there is little point in thinking about it.

But, of course, there is still a logical form to this argument which might be applicable in other cases. It might not work for lifting stones, but maybe it works for something else. Of course, a similar counter-argument to the one above might well apply to other variations of this basic argument structure. Every variation I can imagine would imply some infinite quantity in the physical universe. But my creativity is not endless, so perhaps there is an example I can't think of.

My concern about the logical form of the argument is that it misses a key premise. There are two ways in which we can formulate this premise. The first of them is as below. The second negates the first premise. If the argument were complete, then it might read:

  1. It is logically possible that there is a stone which is logically impossible to lift.
  2. Either God can create a stone which He cannot lift (i.e. which it is logically impossible to lift), or He cannot create a stone which He cannot lift.
  3. If God can create a stone which he cannot lift, then He is not omnipotent.
  4. If God can't create a stone which He cannot lift, then He is not omnipotent.
  5. Therefore, God is not omnipotent.

If that first premise is true, then we see that the third point of the argument is false. If it is logically impossible to lift the stone, then God's inability to lift it is not a barrier to his omnipotence, using the definition of omnipotence above. On the other hand, if we instead reformulate the opening premise to say that it is impossible for there to be a stone which cannot be lifted, then the the fourth point of the argument fails. The definition of omnipotence does not imply that God can create things which are logically impossible, such as square circles (or in this case) infinitely heavy stones. In either case, the conclusion that God is not omnipotent fails. The contradiction thus lies not in the notion of omnipotence, but in the structure of the argument, and its failure to deal with that hidden premise.

This objection doesn't just apply to lifting stones, but to any argument of this sort. Thus I would say that any argument against omnipotence of this nature must fail.

So what is Cowan's defence of this argument? He states that the notion of a being being unable to lift something is not incoherent, and that the notion of a being being able to create something we can't lift is not incoherent. And we know from own own experience that both of these notions are quite credible. But then he makes the mistake of extrapolating from our own experiences as creatures to the creator, or what is logically impossible in general. Again, there is a discontinuity between creature and creator which defeats any extrapolation.

Cowan, in his first essay, references a response to this argument which seems to be similar to my initial objection. He claims that the self-contradiction entailed in the argument is a contradiction in the concept of omnipotence.

His argument rests on the premise,

There are perfectly respectable, even non-self-contradictory predicates, predicates meaningfully abd even truly predicable even of such lowly beings as you and me, predicates which, however, are such that the capacity to have them truly predicated of one logically excludes the capacity to have other similarly non-self-contradictory predicates truly predicated of one.

He justifies this premise by looking at examples which are relevant to the likes of you and me. But then we have the problematic extrapolation to God. Saying that there are predicates which are true for us does not mean that the same thing applies to God. Thus the question is are there such pair of predicates which is applicable to God?

I think that any theist will answer 'No' to this question. The first objection is that Cowan fails to appreciate Aquinas (or Aristotle's) distinction between relational and absolute possibilities. His examples are relational possibilities, that is to say things which are subject to (for example) human power. Yet when we discuss possibility in terms of God's power, we discuss absolute possibilities, that is to say things which are logically possible and logically impossible. Cowan's example is trying to apply relational powers to God. Other examples of this type of argument might neglect the distinction between active and passive powers.

The second objection is whether it is possible to apply predicates to God in the same sense in which we apply them to man. A predicate is sometimes thought of as something which is added to the concept of a referent. For example, in the sentence "Socrates was wise," Socrates is the referent. Socrates is the sort of thing that may or may not be wise. Then we add the predicate "was wise" to tell us something extra about Socrates which goes beyond the definition. But this sort of predication is clearly inapplicable to the notion of God, since it relies on Socrates having accidental properties and God has no accidents. Since God is simple, there is nothing we can add to the notion of God which is not already implied by the notion of God. When we talk about God being wise, we do so analogously. Predication when applied to a created being often refers to that beings accidents. Predication when applied to God can only be a partial description of God.

Thus when we say that "A man cannot lift a stone that is logically possible to lift," we are adding something to our understanding of that man, since the ability or inability to lift stones is accidental to mankind, and even individual people (for example an infant might not be able to lift something which the adult they grow into will be strong enough to do so). But when we extend this to say that "God cannot lift a stone that is logically possible to lift," we are a) attempting to apply the predicate univocally rather than analogously; and b) contradicting ourselves. For predication when used to God describes God's essence, which includes His omnipotence.

Again, there is a discontinuity when we extrapolate from creature to creator.

Thirdly, because of God's simplicity, and that predicates describe the concept of God rather than add something to them, there can be no contradictory predicates in God in the sense that Cowan needs for this argument. Contradictory predicates can apply to creatures, because they can apply to accidental qualities of those creatures, but that is not the case for God. Instead, if there is such a predicate which can be applied to God then there necessarily cannot be self-contradict predicates which can also be applied to God. One or other of the predicates must entail a logical impossibility when extended to the infinite degree we would need for it to challenge God. In the case of the stone, the conception of a stone which is logically impossible to lift is incoherent. For other things, it may be the other horn of the dilemma which entails a logical impossibility. But one of them will always be seen as incoherent after they have been sufficiently modified that they might be applied to God. Cowan has presented an argument which is true for created beings, but he has not shown his first premise is also true of God.

I don't see that Cowan addresses these points in his essays. Thus this argument against God's omnipotence stands unproven.


Professor Stenger admits that theists would not find such arguments convincing.

The philosophers who formulated these disproofs have been careful about defining the terms used, while those who dispute them will generally disagree with those definitions or the way they have been interpreted.

Certainly dispute over the definitions is part of the issue I had with these arguments. But before anyone accuses me of changing the definitions to find loopholes in the arguments, note that when I needed to define a term, I went to a source centuries older than the atheist arguments, either Aristotle or Aquinas (the two of these doctors of the church being in agreement with each other). Indeed the definitions I used were formulated centuries before there were any modern atheists. It is not I who changed the definitions in order to refute the arguments, but the atheists who changed the definitions in order to make the arguments plausible. All I am doing is calling them out on it.

But if there was one issue that connects all these arguments, I would say it is that they fail to appreciate the God of classical theism, and in particular how different God is from any creature. Most of the arguments understand God either through an extrapolation from properties associated with creatures, or by using an ontological definition of God. But we cannot extrapolate across the discontinuity between God and the creatures.

In any case, The impossibility of God is well worth a read for the experienced classical theist. I would not necessarily recommend it to someone who lacks a reasonable understanding of classical theism: you need to know what the authors of the anthology are attempting to criticise in order to be able to properly judge the arguments. I have obviously only discussed a few of the arguments in the anthology here in this post. There are many more. It is a good intellectual exercise to go through the arguments to find the flaws. Some of them are blatant, others a bit more subtle, but I don't think any of the arguments stand up against a properly formulated classical theism.

After this digression looking at what some of the best atheist minds have had to say, it is now time to turn our attention back to what Professor Stenger has to say for himself. So in the next post, I will discuss chapter two of his book, on the illusion of design.

Is God a failed Hypothesis? Part 5: Design Arguments

Reader Comments:

1. Michael Brazier
Posted at 21:40:38 Thursday September 3 2020

Cowan's stones

Cowan's argument can be defeated more simply. A stone that God cannot lift would be a stone that can't be moved by any possible force, since God can create any force which is possible. So any such stone, if God created it, would be logically impossible to lift - and saying that God can't do what is logically impossible is no limit on His omnipotence.

That is, an immovable object and an irresistible force cannot both exist; by definition, the existence of one precludes the possibility of the other. Either one, taken alone, can exist, but the conjunction of both never can. Cowan doesn't distinguish between the two things being possible separately and them being possible together.

2. Nigel Cundy
Posted at 19:11:51 Friday September 4 2020

Cowan's Stones

I agree, it's not a particularly strong argument. With all the arguments, I expect that there are numerous ways of defeating them. On the other hand, brevity has never been my strong point.

3. Cioel
Posted at 00:07:28 Sunday September 6 2020

4. Cioel
Posted at 00:12:50 Sunday September 6 2020

Unhelpful Comments on Reddit Regarding Your Book

Dr Cundy,

What are your thoughts on their brief criticism?

5. Nigel Cundy
Posted at 18:16:33 Sunday September 6 2020

Reddit criticism

Well, at least I'm getting noticed by somebody. But I didn't really see any engagement with my work for me to comment on.

1) I was told that my work was "Pure bullpuckey." But without an explanation for this, it is difficult for me to assess the argument, and if necessary revise or retract what I have written.

2) I apparently am "trying to ride the coattails of science while also attempting to defund and belittle it’s achievements." I am not aware that I have ever tried to defund or belittle the achievements of science. On the contrary, my work assumes that the standard model of physics is (aside from gravity and perhaps some additional particles) correct.

3) Joshua described God as stopping the moon and sun; a topic which is irrelevant to the contents of the book (I stop before discussing Christianity; my work is a defence of classical theism), and easily answered in any case (as someone does later in the thread).

Post Comment:

Some html formatting is supported,such as <b> ... <b> for bold text , < em>... < /em> for italics, and <blockquote> ... </blockquote> for a quotation
All fields are optional
Comments are generally unmoderated, and only represent the views of the person who posted them.
I reserve the right to delete or edit spam messages, obsene language,or personal attacks.
However, that I do not delete such a message does not mean that I approve of the content.
It just means that I am a lazy little bugger who can't be bothered to police his own blog.
Weblinks are only published with moderator approval
Posts with links are only published with moderator approval (provide an email address to allow automatic approval)

What is 4×7+15?