Arguments for the Existence of God

Published Date: 23 Feb 2018

Are any of the arguments for the existence of God successful?

Intro: As an atheist, I find that none of the arguments supporting the existence of God have been successful in persuading me that God does indeed in exist. However since beginning the research for this essay, I have discovered

It if difficult to argue the existence of God without first defining what the word ‘God' means; part of proving the existence of God has to be deciding what or who God actually is. In ‘The Puzzle of God', Vardy sites Plato's description of God as kind of ‘moulder of clay' (Vardy, 1999: 17). God did not create the matter which makes up the universe, he merely utilised it to create planets and animals - He moulded the matter to his own ends, made special through implanting his ‘mind' and ‘soul' into it, which the dead matter alone would not have possessed. This concept does not strictly fit the Judaeo-Christian belief that God created the world out of nothing (Genesis 1:1, New International Version) so it is already apparent that the concept of God cannot be discussed in simple terms. Indeed the major monotheistic religions, Islam, Judaism and Christianity, whilst all believing in one God, are varying in their views on what /who God is (Davies, 2004: 1). However, for the purposes of this essay, I shall focus on the Christian beliefs about God, as the major philosophers of religion, St Anselm, Thomas Aquinas, Descartes et cetera have debated the existence of God through a Christian lense. The God of Christianity is often portrayed anthropomorphically - he is an old man with a grey beard in many children's eyes - and it is therefore hard not to imbue him with human traits such as anger and joy and was also made flesh through the living man, Jesus (Martin, 1959: 2/3).

Mackie - Descartes' idea of god.

Pascal's wager - sensible! However, can discount this as acting as you have  faith is different to actually having one. Let's look at some agreed upon arguments supporting the existence of God.

1. First para: It is very difficult to decide if God exists, without first defining what the word or concept of ‘God' is/means. In this essay, I will focus on the monotheistic faiths' definition of God, if indeed there is an agreed definition. In ‘An Introduction to the Philosophy of Religion', Brian Davies decided to focus on ‘Classical theism' and ‘theistic personalisation'.
2. Mention Pascal's Wager - is it a good reason for believing in God, in a way it's the one I most identify with, in a grave yard at night I'm scared of ghosts & if I were hanging off a cliff I might pray….
3. Firstly - design (teleological) argument, when considering the Q, I look around and use my senses, empirical evidence to decide. This strongly influences me and makes me reconsider, however, if you believe in natural selection, that can dispel the ‘watchmaker' (1000)
4. Criticisms of the TA and points I agree with
5. Leads onto how did the word start - cosmological - this is the most convincing argument to me, as my world is finite and I have often wondered what we are expanding into. Does make sense in my mind that (1000) Aristotle's first cause (see revision guide).
6. Criticism of CA and points I agree with in them
7. Ontological argument - another argument which I find the least convincing, despite it described as being the most 'logical' argument by its supporters. When I read that this was a logical argument, I became immediately more interested as I think of myself as  logical person, however, it disappoints. The whole argument hinges on the definition of God. If the definition of God is disputed so how can the definition of God prove His existence?
8. Criticisms  of the OA - what I agree with in them
9. Moral argument (Kant) and criticisms. Does moral order need to be underwritten by God? Reference radio 4 prog?

Design argument

Arguments for design

A common human response to the world around us is amazement. Two things in particular are sources of fascination - the stars in the sky and life. The grandeur of a star-filled night, the vastness of the universe, inspires a sense of awe. The complexity and intricacy of living creatures fills us with wonder. As philosophers, we should first of all be amazed that we can understand the world at all. It could have been a complete shambles, nothing constant, no laws of nature, no means by which our reason could try to explain it. But what we find is order, constancy, predictability throughout, and in living creatures, different parts working together and the creature as a whole fitting neatly into its environment.

Life

When we talk about parts of a living creature, we often refer to their purpose. The heart is for pumping blood; the eye is for seeing; and so on. In fact, this is central to understanding the organ in question. You don't really know what an eye is unless you know that it is the organ of sight. And we want parts of the eye in terms of their contribution to the purpose of the eye. So the lens focuses light onto the retina, the muscles attached to the lens change its thickness so that it can focus light onto the retina, and so on. Without this bit (the lens) or that bit (the retina), the eye wouldn't work properly.

The way in which living things work, which requires a huge coordination of lots of tiny bits, each doing their specific job, is amazingly complex. [Margin: Similarly, the way living creatures interact in an ecosystem, each filling an ecological ‘niche', is highly complex. Remove one creature and the ecosystem can become unstable and start to break down.] This coordination, the detail and intricacy of interrelations between parts, suggests planning - a plan that follows a purpose (of making a living creature, making an organ that enables the creature to see, etc.). Acting on a plan guided by a purpose is design. It's as if someone had it in mind that the eye should see, and put the bits together to ensure that it could. The way living creatures are suggests that they are designed - designed to be alive, with organs designed to keep them alive.

If living creatures are designed, then as a matter of definition, there must be a designer. You can't have design without a designer. This is the next step in an argument for the existence of God, which we look at in the separate handout on ‘Arguments from design'. (The Greek word for this idea of ‘purpose' or ‘end' or ‘what is aimed at' is telos. Arguments for the existence of God that invoke purpose or design are therefore also called teleological arguments.)

Evolution by natural selection

Is the fact that we are amazingly complex, and our organs and many parts serve the purpose of keeping us alive enough for us to say that living creatures are designed? The appearance of design in nature clearly requires some explanation. But what is the best explanation?

Darwin's theory of evolution by natural selection provides an excellent account of how the appearance of design could come about without anything being designed. Millions of alterations randomly take place. Most disappear without a trace. But some trait that coincidentally helps an creature to survive and reproduce, to function well, slowly spreads, because that creature and its descendents reproduce more. So more and more creatures end up with it. It's not that the feature is ‘selected' in order for the creature to live better and so reproduce more. Instead, the feature simply causes the creature to reproduce more, so its descendents also have that feature and they reproduce more and so on. What appears to be designed is actually just evidence of good functioning.

One very small change is followed by another. Over time, this can lead to great complexity. In the end, then, creatures appear to be designed when they are in fact the product of coincidence. So we don't need to say that living things are actually designed (which would require the existence of a designer). This is a better explanation because it is simpler. We aren't inferring the existence of something we can't be sure exists.

The ‘fine tuning' argument

Darwinism is sometimes thought to eliminate the question of design in nature. But we can ask ‘how is evolution by natural selection possible?'. It didn't have to be possible - perhaps the universe could have been organized in such a way that evolution and life would be impossible. So the appearance of design needs a further explanation. For example, perhaps God set up the universe so that life evolves by natural selection.

This argument has been given support by recent work in cosmology - the study of the ‘cosmos' or universe as a whole. Cosmologists have argued that the conditions needed for life to come into existence are incredibly improbably. As far as we understand, life needs planets; and planets need stars. But the universe needn't have contained stars. In fact, if anything about the beginnings of the universe (the Big Bang) or the laws of nature were different by the smallest amount, stars wouldn't exist. For example, the Big Bang was an explosion of matter-energy. Logically speaking, it could have been bigger, it could have been smaller - either more or less matter or more or less force of explosion. But for stars to be able to form, the initial strength of the explosion in the Big Bang had to be precise to one part in 1060 - it couldn't vary by more than 0.00000000000000000000000000000000000000000000000000000000001 percent. That's as precise as hitting a one-inch target at the other side of the observable universe! And that's just for stars to form. For life to form on planets is even more improbable, because so many more laws are involved. Because it is so unlikely, the fact that everything is exactly adjusted so life can exist seems a staggering coincidence.

What could explain this? Science can't tell us why the Big Bang was exactly the size it was or why the laws of nature are the way they are. It can only tell us that this is how it is. One obvious explanation, then, is that the Big Bang, the properties of matter-energy, the laws of nature, were all designed to allow life to evolve. If they were designed, then instead of it being a massive coincidence that life could evolve, it becomes inevitable.

Paley and design-like properties

A famous argument for design was given by William Paley in his Natural Theology, Ch. 1-3. He argues that if I found a stone lying in a field, and wondered how it came to be there, I might rightly think that, for all I knew, it had always been there. However, if I found a watch lying on the ground, I wouldn't feel the same answer was satisfactory. Examining it closely, I would infer it had a designer.

Now we know that watches have designers. But what is it about a watch itself that leads us to think it must have a designer? Paley spends a considerable time exploring this inference, and whether it is valid in the case of the watch. For example, would it undermine the inference if the watch sometimes went wrong, or if I'd never seen a watch being made? He is trying to identify exactly what it is about a watch that allows us to infer a designer. After all, in the case of a watch, this does seem a good inference. Watches don't just ‘happen'. What properties of a watch are direct evidence of design?

Paley identifies the property of having an organization of parts put together for a purpose as crucial. It is from this that we infer the watch has a designer - even if we know nothing about watchmakers. He then argues that we can make exactly the same argument in the case of natural things that exhibit that property. [Margin: It is often thought that Paley argues from analogy; but he is not arguing that natural things are like watches. He is arguing that watches have a property which supports the inference of a designer, and then arguing that natural things have exactly that property as well.] Natural things have the same property, so they too have a designer.

An objection

Throughout the argument, Paley is relying on the idea that the sorts of properties he takes as evidence of design - in the case of the watch and of nature - cannot be produced by natural means, and so must be the result of a mind. In the case of the watch, this seems right - a watch isn't the kind of thing nature produces. So if we found a watch in a field, we would rightly wonder about its origin. However, natural things are precisely the sort of thing that nature does produce. We can't, then, argue that natural things cannot be produced by natural means, so must have been designed by a mind.

So what is the difference between natural things and watches? There is no question that natural things have design-like properties. Paley has established this. The difficulty is that unlike watches, natural things don't show evidence of being manufactured artefacts. In this different context, their design-like properties aren't clearly good evidence for actually having been designed. Although we are making the same inference from design-like properties to a designer, the argument doesn't have the same force in the case of natural things. And as in our earlier discussion, we can appeal to Darwinism to show that nature can produce design-like properties (though not manufactured artefacts).

‘Intelligent design'

In the last twenty years, some thinkers have become dissatisfied with explanations of the complexity of living creatures in terms of evolution by natural selection. They have been struck by the ‘irreducible complexity' of the systems and organs of living creatures. Michael Behe, a biochemist, defines irreducible complexity as

a single system which is composed of several interacting parts that contribute to the basic function, and where the removal of any one of these parts causes the system to effectively cease functioning'. (‘Molecular machines')

This property is precisely the one that Paley has argued is the basis for inferring a designer.

As an example of such a situation, Behe describes the many parts (over 40) that work together to move the ‘tail' that propels a certain bacterium. Behe argues that evolution couldn't produce such an organization of parts. The reason is that, as we saw, evolution works by making small changes, accidentally, and one at a time. But until all the pieces are in place together, the tail wouldn't work. It's all or nothing. But evolution is bit by bit.

Like Paley, Behe argues that irreducible complexity is direct evidence of design. If a system won't work at all until all its parts are in place, then this suggests someone planned and organized the parts.

Objections

However, many evolutionary biologists reject this conclusion. First, Behe's argument assumes that each part in a system has always been that part in that system. But this isn't true in evolution. It often happens that a system, or its parts, having evolved to do one thing are ‘co-opted' into doing something else. Some of the parts that move the bacterium's tail, work as a kind of pump if taken alone. They may have had nothing at all to do with movement when they first evolved. They could have evolved as a pump, and then later on, some further accidental change meant they joined with some new part to move a tail.

Second, features that are initially minor improvements can become essential. Take lungs - very complex and without which we wouldn't survive. But they started out as relatively unimportant air bladders in fish (they help fish not sink to the sea floor, but not all fish have them, e.g. sharks do not). They acquired a new function when fish made brief forays onto land, now operating to supply the fish with oxygen as well. Over time, developments in the air bladder served this new purpose, allowing for longer and longer trips out of water. Eventually they became lungs, and the fish ceased to be fish. Lungs didn't have to evolve all at once. So we can argue that evolution by natural selection can account for irreducible complexity.

Arguments from design

Arguments from design start from this evidence of design and infer the existence of a designer, a mind that can order things for a purpose. The most famous of these is the argument from analogy.

The argument from analogy

In Dialogues on Natural Religion, Part II, David Hume expresses the argument like this

The curious adapting of means to ends, through all nature, resembles exactly, though it much exceeds, the productions of human contrivance; of human design, thought, wisdom, and intelligence. Since, therefore, the effects resemble each other, we are led to infer, by all the rules of analogy, that the causes also resemble; and that the Author of Nature is somewhat similar to the mind of man, though possessed of much larger faculties, proportioned to the grandeur of the work which he has executed.

Hume is saying that nature is like human inventions in the way it displays purpose (the adaptation of means to ends, e.g. the arrangement of the parts of the eye to see, of the heart to pump blood), so it must have a similar cause to human inventions, viz. a mind that intended to create such design. Similar effects have similar causes.

Objections

However, Hume argues strongly against the analogy. First, he questions its strength. A watch is a typical example of something designed and made by humans. But living creatures aren't really like watches in all sorts of ways. For example, watches aren't alive and they don't reproduce. So the ‘effects' - watches, living creatures - aren't all that similar, so we can't infer a similar cause. Likewise, the universe is not at all like a watch. So again, because the effects aren't very like, we can't infer similar causes.

Second, even if the analogy between effects was better, inferring a similar cause would be dubious. Human beings are a fairly recent species living on a small planet on one of billions of galaxies. We can't reliably generalise from our very limited and finite experience to the cause of the universe as a whole. As Hume says, ‘why select so minute, so weak, so bounded a principle as the reason and design of animals is found to be upon this planet' as a model for something that could set laws of nature?!

Third, there could be other explanations of apparent design. With life, this is evolution. We don't know what might explain the universe, but then, that's the situation we were in about life before Darwin developed his theory. Hume suggests the idea that if the universe is infinitely old, then over time, all possible combinations of matter will occur randomly. This suggestion isn't very good, because we know that the universe began around 13.8 billion years ago, and we know that matter doesn't organize itself randomly, but follows very particular laws of nature. But Hume's point is that if there are different explanations of the apparent design of the universe, then we can't infer that the cause is a designer.

Is the designer God?

The argument from design is intended as an argument for the existence of God. However, as well as attacking the analogy, Hume also points out that even if we could infer the existence of a designer of the universe, it is an extra step to argue that the designer is God. And, because we are relying on analogy, this extra step also faces difficulties.

Let's take the analogy between human inventions and the universe further. First, we should note that, in the human case, the designer is not always also the creator. Someone who designs a car may not also build it. So we can't infer that the designer of the universe also created the universe. But God is said to be the creator of the universe; so we can't infer that the designer is God.

Second, the scale of the design reflects the powers of the designer. Watches aren't infinite, and neither are the minds that make them. But the universe isn't infinite either. So we can't infer that the designer is infinite, only that whoever designed the universe has sufficient power and intelligence to do that. But God is said to be infinite.

Third, we think that the quality of what is designed reflects the abilities of the designer. Designers need to be trained, and at first their designs will be poor and could be improved. We can argue that, if the purpose of the universe was life, this universe shows examples of poor design, e.g. volcanoes and tsunamis that wipe out life. [Margin: This idea is discussed further in The problem of evil (p. xxx).]  Perhaps we should infer that the designer of this universe was not fully skilled, but made mistakes. But God is said not to make mistakes.

Some of Hume's points can be debated. But the overall message is clear: If we rest the argument from design completely on analogy, then the argument faces many problems. What philosophers have done since Hume is to remove the appeal to analogy.

Swinburne's argument

We can do this by using the considerations about probability. Cosmology supports the view that it is hugely improbable that the universe would have the right properties for life to evolve. But if God exists, we can explain this. So it is more probable that God exists and designed the universe for life than that the universe just randomly happened to have the right features for life. This is an inductive argument from probability for the existence of God.

The argument only works if God is the only satisfactory way that we can explain the fact that the universe allows life to evolve. In other words, we need to ask whether God is the best explanation for this fact. For example, could we not give a scientific explanation? In The Coherence of Theism (Ch. 8), Richard Swinburne argues not. Science can't offer any satisfactory explanation, because science can't provide us with the right sort of answer to why the universe has the laws it has or the exact quantity of matter it has. Science must assume the laws of nature in order to provide any explanations at all. It can't say where they come from or why they are the way they are, because all scientific explanations presuppose laws.

For example, science explains why water boils when you heat it in terms of the effect on heat on the properties of molecules. It explains these effects and these properties in terms of other laws and properties, atomic and sub-atomic ones. Some further explanation of these may be possible, but again, it will suppose other laws and properties. So at root, scientific laws are ‘brute' - they have no explanation unless we can find some other kind of explanation for them.

We use another type of explanation all the time, viz. ‘personal explanation'. We explain the products of human activity - this book, these sentences - in terms of a person. I'm writing things I intend to write. This sort of explanation explains an object or an event in terms of a person and their purposes. The hypothesis that God exists and intended to life to evolve provides a personal explanation for why the universe is such that life can evolve.

Best explanation

However, we saw that Hume objected that even if you can show that the universe has a designer, you can't show that the designer is God, as we normally think of God? For example, this argument doesn't show that there is only one cause of the universe; nor does it show that that cause is perfect, omniscient, omnipotent, or cares about people. The argument only needs ‘God' to be able to design the universe (and perhaps, put that design into effect). It doesn't say anything else about God.

Swinburne's response is to accept this objection. The argument so far is only evidence for a designer, not evidence for the traditional theistic conception of God. However, he argues, the argument is about what is the best explanation for design; and God as we usually think of him remains the best explanation.

Swinburne says an explanation is good ‘when the explanatory hypothesis [in this case, the existence of God and his intention for the universe to contain life] is simple and leads us with some probability to expect the data which we would not otherwise expect.' ‘Simplicity' means not invoking more different kinds of thing than you need to; and not giving them more or more complex properties than they need for the explanation to work.

Simplicity requires that we shouldn't suppose that two possible causes exist when only one will do. Supposing there is more than one cause of the universe is a worse explanation, because it is not as simple. It is also simpler to suppose that the cause of the universe is itself uncaused, or we have a problem of regress. It is also simpler to suppose that God has infinite power and intelligence, or we would have to explain why God had just the amount of power and intelligence he has (enough to create the universe, but no more), i.e. what limits God's power and intelligence.

(Swinburne adds infinite goodness to the properties of God, but we can question this - why does God need to be good in order to create the universe?

The limits of explanation

If we explain design in terms of God, now we have to ask ‘What explains God?' and this seems to be an even more puzzling question than ‘What explains scientific laws?'. So from not being able to explain design in the universe, we end up not being able to explain something else. This is not progress.

Swinburne responds that it is progress, and that we do something similar all the time in science. Science will introduce an entity - like sub-atomic particles - in order to explain something, e.g. explosions in a nuclear accelerator. However, these new entities now need explaining, and scientists don't yet know how to explain them. This is absolutely normal, and has happened repeatedly throughout the history of science. It is progress, because we have explained one more thing. So we can still say that God is a good explanation for scientific laws even if we can't explain God.

But if we will always have something we can't explain, why invoke God? Why not just say we can't explain scientific laws? Because scientific laws leave fewer things unexplained, and we should explain as much as we can. This is a principle of science and philosophy. If you give up on this, you give up on pursuing these forms of thought.

Does the universe need explaining?

But do we need any explanation for why the universe appears designed? Some things that appear to be coincidence are in fact inevitable, e.g. winning the lottery: it is very unlikely that you will win, but it is inevitable that someone will win. For whoever wins, that they won is a huge coincidence; but we don't need any special explanation for it (such as ‘someone intended them to win, and rigged the lottery').

Suppose, then, that instead of just this universe, there are or have been millions of universes. Each had different scientific laws, and in most cases, the laws didn't allow the universe to continue to exist - as soon as it began, it ended. Others existed, but there was no life. It was inevitable, we might think, that given all the possible variations in scientific laws, a universe such as ours would exist, and therefore so would life. It doesn't need any special explanation - it had to happen.

But why ours? Well, it had to be ours because we wouldn't be here if it wasn't! Given that life does exist in it, this universe has to have the right scientific laws for life to exist. If it didn't, life wouldn't exist in it. There is nothing special about this universe, except that it has the right laws; just like there is nothing special about the ticket that wins the lottery. [Margin: Explain the argument that the appearance of design in the universe needs no special explanation.]

But we can object that this response assumes the existence of huge numbers of other universes, which are completely inaccessible to us, and for which we have (virtually) no other evidence. Why should we assume that? The existence of God, by contrast, Swinburne argues, is simpler (just one God, not millions of universes) and is also supported by other evidence, e.g. miracles and religious experience. So the existence of just one universe, designed by God, is a better explanation.

We can object, however, that we also have evidence against the existence of God, viz. the problem of evil. At least we don't have evidence against the existence of other universes.

Cosmological argument

The question at the heart of the cosmological argument is ‘why does anything exist? why something rather than nothing?'. The argument is that unless God exists, this question is unanswerable. There are different forms of the argument. Two central ones are the Kalam argument and the argument from contingent existence. They are usually presented as deductive arguments; an inductive variation is given by Richard Swinburne.

The Kalam argument

The Kalam argument observes that

1. of anything that begins to exist, you can ask what caused it. For example, what caused me (my birth)? In a sense, my parents. But then, we can repeat the question: ‘what caused my parents?' And so on. We can go back to the beginning of the universe,  and then ask ‘what caused the universe?'. If
2. the universe began to exist, then
3. it must have a cause of its existence. Something can't come out of nothing.
4. What we need is something that causes things to exist, but the existence of which isn't caused itself.
5. Only God could be such a thing.

There are three key issues that need to be addressed to defend the argument. First, is the causal principle, that everything that begins to exist has a cause, correct? Second, does the universe have a beginning? Third, must the explanation be God? We will leave this third question to the very end of this handout.

The causal principle

Must every event have a cause? David Hume famously argued that we cannot know this. It is not an analytic truth (by contrast, ‘every effect has a cause' is an analytic truth; but is every event an effect?). ‘Something cannot come out of nothing' is also not analytic. And Hume argued that synthetic truths are known a posteriori, through experience. And although our experience is that everything so far has a cause, can this principle can be applied to the beginning of the universe?

First, the beginnings of universes is not something we have any experience of. Second, the beginning of the universe is not an event like events that happen within the universe. It doesn't take place in space or time, since both come into existence with the universe. Even if everything within the universe has a cause, that doesn't mean that the universe as a whole does. We cannot apply principles we have developed for events within the universe to the universe as a whole. Bertrand Russell famously put it: ‘the universe is just there, and that's all'.

Does the universe have a beginning?

Rather than challenge the causal principle, we can reject the idea that the universe has a beginning at all. Because time came into existence with the universe, the universe didn't ‘happen' at a time, so in a sense, it has no beginning. We can reply that, even if this is true, science suggests the universe has a finite past (it is about 15 billion years old). Whatever has a finite past must have a cause of its existence. In the case of the universe, that cause can't exist in time if time didn't exist before the universe. But that doesn't mean there was no cause, only that the cause must exist outside time. Which God does, according to many theists.

Alternatively, even if this universe has a beginning, perhaps it was caused by a previous (or another) universe, and so on, infinitely. In other words, rather than infer that God exists, we may think there is just an infinite regress of causes. Something has always existed.

It is, however, difficult to imagine what infinity is; it is not, for instance, simply a ‘very long time'. It is very different from a ‘very long time' - it means, quite literally, that there was no beginning, ever. Because the universe exists, this response claims that an actual infinity - something that is in fact infinite - exists. This is quite different from talking about the idea of infinity. The idea of infinity makes sense; but does it make sense to think that something infinite exists?

Infinity

For example, the universe gets older as time passes. But this couldn't happen if the universe was infinitely old, because you cannot add any number to infinity and get a bigger number: ¥ + 1 = ¥. So if the universe is infinitely old, it is not getting any older as time passes! Or again, to have reached the present, an infinite amount of time would need to have passed. But it is not possible for an infinite amount of time to have passed, since infinity is not an amount. So if the universe was infinitely old, it could never have reached the present.

Given that science tells us the (this) universe has a beginning, this discussion of something always existing means that we must think of preceding universes. But given that the beginning of this universe was also the beginning of time as we know, we may wonder what sense to make of talking about anything existing before this universe. We should not talk about an infinity of time, therefore, but an infinite series of causes (some operating outside the time of this universe).

But the puzzles arise for an infinite series of causes, too. Each new cause doesn't add one more cause to the series, since ¥ + 1 = ¥. And we would never have reached the point in the series at which we are now if it were an infinite series.

We noted that the question at the heart of the cosmological argument is ‘why something rather than nothing?'. If we have an infinite series of causes, although each cause can be explained in terms of the previous cause, we may wonder what explains the whole series. If we say something exists because something has always existed, we still haven't answered the question why anything exists at all. This takes us to the next form of cosmological argument.

The argument from contingent existence

This version of the cosmological argument, defended by Frederick Copleston in a radio debate with Bertrand Russell, emphasises the need to explain what exists.

1. Things in the universe exist contingently, they might not have existed or they might stop existing.
2. Something that exists contingently has (and needs) an explanation of why it exists; after all, it is not inevitable.
3. This explanation may be provided by the existence of some other contingent being (as in the example of me and my parents). But then we must explain these other contingent beings.
4. To repeat this ad infinitum is no explanation of why anything exists at all.
5. So what explains why contingent beings exist at all can only be a non-contingent being. A non-contingent being is one that cannot not exist, i.e. it exists necessarily, and doesn't need some further explanation for why it exists.
6. This necessary being is God.

Objections

Russell accepts that of any particular thing in the universe, we need an explanation of why it exists, which science can give us. But it is a mistake to think that we can apply this idea to the universe itself. A form of explanation developed for the parts of the universe needn't apply to the universe as a whole.

However, we can reply that the universe is itself a contingent being - if every part of the universe ceased to exist, so would the universe. So as a contingent being, the universe is like its parts. What is wrong with the principle that all contingent beings require an explanation for their existence?

A second objection is that although, as philosophers and scientists, we should look for explanations of contingent beings, we cannot know that in fact, every contingent being has such an explanation. Without this, the argument fails as a deduction. However, this objection can be avoided if we give up the idea that the cosmological argument is deductive, and claim it is an inference to the best explanation instead (see below).

A third objection attacks the conclusion. It is not God but, matter/energy (in some form) that is a necessary being. A fundamental law of physics is the conservation of energy: the total amount of matter/energy in the universe remains constant, it cannot be increased or decreased. If a version of this law applied even at the beginning and end of universes, then matter/energy is a necessary being. However, we have no reason to believe that this law does apply at the beginning (and possibly the end) of the universe. The Big Bang theory suggests that matter/energy was created, along with time and space, i.e. the universe came into existence - so it is contingent.

Swinburne: an inductive argument

Richard Swinburne claims that the cosmological argument is better understood as an inference to the best explanation (The Coherence of Theism). God's existence isn't logically proven, but it is probable, given the premises. Considered on its own, the claim ‘God exists' is very improbable, says Swinburne. But in light of the cosmological argument, it becomes more probable, because God's existence is the best explanation for why the universe exists.

An inductive argument for God's existence needs to take into account all the evidence, both for and against. Swinburne does not defend God's existence on the basis of the cosmological argument alone. He combines it with other arguments. There is also a similar version of the argument from design, viz. that God is the best explanation of the order and purpose that we find. We can add the argument from religious experience and an argument from miracles. Each work the same way: the existence of God is the best explanation for these phenomena. When we put all these arguments together, he claims, it becomes more probable that God exists than that God doesn't.

Swinburne's cosmological argument

If we look over the two cosmological arguments above, it is apparent that we can't deduce God's existence. But the premises are plausible, and the inferences are intuitive. So although it is not an analytic truth that everything that begins to exist has a cause, it is extremely probable - our experience supports it. And the theory of the Big Bang and the problems with the infinite existence make it more plausible that the universe (or matter/energy) has not existed without beginning, but came into existence. If we reject God as an explanation for the existence of the universe run into problems - if not God, then what?

The second part of Swinburne's argument is that we have reason to believe that no other explanation of the universe will be satisfactory. For instance, any scientific explanation must already assume that something exists, and that whatever exists is governed by scientific laws. If we explain this universe in terms of another universe, we then have to explain the existence of that universe. And science can't explain scientific laws - where they come from or why they are the way they are, because all scientific explanations presuppose laws. Scientific laws are ‘brute' - they have no explanation unless we can find some other kind of explanation for them.

Explaining the existence of the universe in terms of God doesn't suffer this problem, because it is not a scientific explanation, but a ‘personal' one. We explain the products of human activity - this book, these sentences - in terms of a person. I'm writing things I intend to write. This sort of explanation explains an object or an event in terms of a person and their purposes. The hypothesis that God exists and intended to create the universe (including its laws) provides a personal explanation for the existence of the universe. So, Swinburne argues, it is probable that God exists and caused the beginning of the universe.

Best explanation

But is this the best explanation, or does it face as many difficulties as scientific explanation? Does invoking God's existence just make us more puzzled?

First, does the argument support the existence of God, as we normally think of God? It doesn't show that there is only one cause of the universe; nor does it show that that cause is perfect, omniscient, omnipotent, or cares about people. The cosmological argument only needs ‘God' to be able to create the universe. It doesn't say anything else about God.

However, a good explanation will be simpler than its rivals. ‘Simplicity' means not invoking more different kinds of thing than you need to; and attributing only those properties that they need for the explanation to work. So simplicity requires that we shouldn't suppose that two possible causes exist when only one will do. Supposing there is more than one cause of the universe is a worse explanation, because it is not as simple. It is also simpler to suppose that the cause of the universe is itself uncaused, or we have a problem of regress. It is also simpler to suppose that God has infinite power and intelligence, or we would have to explain why God had just the amount of power and intelligence he has (enough to create the universe, but no more), i.e. what limits God's power and intelligence.

(Swinburne adds infinite goodness to the properties of God, but we can question this - why does God need to be good in order to create the universe? This objection becomes more pressing in light of the problem of evil.)

The limits of explanation

We can object that Swinburne has not demonstrated that God is the best explanation for the existence of the universe, because we are left with the question ‘What explains God?' and this seems to be an even more puzzling question than ‘What explains scientific laws?'.

Swinburne responds that science will introduce an entity - such as a type of sub-atomic particle - in order to explain something, even though that entity needs explaining itself, and scientists don't yet know how to explain it. So we can still say that God is a good explanation for scientific laws even if we can't explain God.

But if we will always have something we can't explain, why invoke God? Why not just say we can't explain scientific laws? Russell, for instance, rejects the idea of trying to give an explanation for the universe at all.  But Swinburne responds that it is a principle of science and philosophy. If you give up on this, you give up on pursuing these forms of thought. If we invoke God, we can explain scientific laws and the existence of the universe, and we should explain as much as we can.

Ontological argument

We are starting with the thought that nothing could be greater than God. Another way this thought has been expressed is that God is perfect. Augustine says that to think of God is to ‘attempt to conceive something than which nothing more excellent or sublime exists'. Some philosophers claim that God is the most perfect being that could (not just does) exist. St Anselm and Descartes both famously presented an ontological argument for the existence of God. (The word ‘ontological' comes from ‘ontology', the study of (-ology) of what exists or ‘being' (ont).) Their versions of the argument are slightly different, but they both argue that we can deduce the existence of God from the idea of God. Just from thinking about what God is, we can conclude that God must exist. Because it doesn't depend on experience in any way, the ontological argument is a priori.

St Anselm's version of the ontological argument

Anselm's argument relies on ‘conceivability':

1. By definition, God is a being greater than which cannot be conceived.
2. I can conceive of such a being.
3. It is greater to exist than not to exist.
4. Therefore, God must exist.

The idea of God as the most perfect possible being has a long history. And perfection has also been connected to reality: what is perfect is more real than what is not. Anselm's argument makes use of both these ideas.

Anselm starts from a definition of God - if we could think of something that was greater than the being we called God, then surely this greater thing would in fact be God. But this is nonsense - God being greater than God. The first being isn't God at all. We cannot conceive of anything being greater than God - if we think we can, we're not thinking of God.

The second premise says that this idea - a being greater than which we cannot conceive - is coherent. Now, if we think of two beings, one that exists and one that doesn't, the one that actually exists is greater - being real is greater than being fictional! So if God didn't exist, we could think of a greater being than God. But we've said that's impossible; so God exists.

Objections

Gaunilo and the perfect island

Anselm received an immediate reply from a monk named Gaunilo: you could prove anything perfect must exist by this argument! I can conceive of the perfect island, greater than which cannot be conceived. And so such an island must exist, because it would be less great if it didn't. But this is ridiculous, so the ontological argument must be flawed. You can't infer the existence of something, Gaunilo argues, from the idea of its being perfect.

Anselm replied that the ontological argument works only for God, because the relation between God and greatness or perfection is unique. An island wouldn't cease to be what it is - an island - if it wasn't perfect; of course, it wouldn't then be a perfect island. But islands aren't perfect by definition; perfection is something an island can have or not have. It is an ‘accidental' not an ‘essential' property of islands. It's perfectly coherent to think of an island that isn't perfect.

(An essential property is one that something must have to be the thing that it is. Islands must be areas of land surrounded by water. Can we say that ‘perfect islands' are islands that are essentially perfect? Not convincingly, because perfect islands aren't a different kind of thing from islands, but a type of island. So they are still only essentially islands, and accidentally perfect.)

By contrast, God, argues Anselm, must be the greatest conceivable being - God wouldn't be God if there was some being even greater than God - it's incoherent to think of God as imperfect. Being the greatest conceivable being is an essential property of God. But then because it is better to exist than not, existence is an essential property of God. So to be the greatest conceivable being, God must exist.

Hume on necessary existence

Notice that this conclusion is more than ‘God does exist'; it claims God must exist - God's existence is necessary. That isn't true of you or me or islands - we can exist or not, we come into existence and cease to exist. Our existence is contingent. (The existence of a being is contingent if it could be true or false that that being exists, e.g. it could now exist, but later cease to exist. The existence of a being is necessary if it cannot come into or go out of existence; it is necessarily true that it exists (or doesn't).) The ontological argument only works for God, says Anselm, because only God's existence could be necessary.

Hume argued that the idea of ‘necessary existence' was meaningless (Dialogues on Natural Religion, § IX). To understand his claim, we need to understand what Hume thought about knowledge. Hume argues that we can have knowledge of just two sorts of thing: the relations between ideas and matters of fact (Enquiry Concerning Human Understanding, § IV) His distinction was developed by later philosophers, and is now understood in terms two distinctions: analytic/synthetic and a priori/a posteriori.

Matters of fact we establish through sense experience, and are a matter of evidence and probability. The ontological argument doesn't rely on sense experience, but on pure reasoning. So the argument, and its conclusion that God exists, are a priori. But the only claims that can be known a priori are ‘relations of ideas'. These are ‘demonstrable', i.e. provable, not a matter of probability, but certain. Take the claim ‘all vixens are female'. What is a vixen? By definition, it is a female fox. So ‘all vixens are female' means ‘all female foxes are female'. To deny this is to contradict oneself; if not all female foxes are female, then some female foxes are not female. But how can a female fox not be female?!

If ‘God exists' is a priori, then we shouldn't be able to deny it without contradicting ourselves: ‘Nothing is demonstrable, unless the contrary is a contradiction', Hume says. But, he goes on,

Whatever we conceive as existent, we can also conceive as non-existent. There is no being, therefore, whose non-existence implies a contradiction. Consequently there is no Being whose existence is demonstrable.

So Hume argues that God does not possess existence essentially - it is possible to conceive of God not existing (and still be thinking of God). So God does not exist necessarily. And so the ontological argument for God's existence fails.

However, Hume assumes that all demonstrable truths must be analytic. Rationalists would argue that there are synthetic a priori truths that are demonstrable as well, and the claim that ‘God exists' could be one of these.

Descartes and perfection

We saw St Anselm's version of the ontological argument relied on the inconceivability of anything greater than God. Descartes' version of the argument relies on perfection alone, not conceivability:

It is certain that I… find the idea of a God in my consciousness, that is the idea of a being supremely perfect: and I know with… clearness and distinctness that an [actual and] eternal existence pertains to his nature… existence can no more be separated from the essence of God, than the idea of a mountain from that of a valley… it is not less impossible to conceive a God, that is, a being supremely perfect, to whom existence is wanting, or who is devoid of a certain perfection, than to conceive a mountain without a valley. (Meditation V)

Descartes' argument is this:

1. I have the idea of God;
2. God is a supremely perfect being;
3. Existence is a perfection;
4. Therefore, God must exist.

Development

We can object that there is a gap between the idea that God exists eternally and God actually existing eternally. Descartes is aware of this, and objects to himself:

just as it does not follow that there is any mountain in the world merely because I conceive a mountain with a valley, so likewise, though I conceive God as existing, it does not seem to follow on that account that God exists.

But, he replies,

the cases are not analogous… it does not follow that there is any mountain or valley in existence, but simply that the mountain or valley, whether they do or do not exist, are inseparable from each other; whereas, on the other hand, because I cannot conceive God unless as existing, it follows that existence is inseparable from him, and therefore that he really exists.

Descartes is arguing that the analogy is not between mountains and existence and God and existence; but between mountains and valleys and God and existence. The idea of existence is no part of the idea of a mountain. But just as the idea of a valley is implied by the idea of a mountain, so the idea of existence is part of the idea of God. And so, as he says, I can't think of God without thinking that God exists.

But what does this show? Just because I can't think of God not existing, does that have any relevance to whether or not God exists? Absolutely. The bounds of our thought are, at least on some occasions, indications of what is possible. This isn't because our thought creates or influences reality, but because thought reveals reality. And so, Descartes argues, the necessary connection between God and existence isn't something I've come up with, it is something I discover:

the necessity which lies in the thing itself, that is, the necessity of the existence of God, determines me to think in this way: for it is not in my power to conceive a God without existence.

There is a conceptual connection between the concept of God and God's existence, and this entails that God's must exist.

Objections to Descartes

Gassendi objects that existence is not part of the idea of God as a supremely perfect being. Can't I form the idea of a God who does not exist? (This is similar to Hume's objection.) Descartes replies by drawing claiming, with St Thomas Aquinas, that divine perfections all entail each other. Because our minds are finite, we normally think of the divine perfections - omnipotence, omniscience, necessary existence, etc. - separately and ‘hence may not immediately notice the necessity of their being joined together'. But if we reflect carefully, we shall discover that we cannot conceive any one of the other attributes while excluding necessary existence from it. For example, in order for God to be omnipotent, God must not depend on anything else, and so must not depend on anything else to exist.

However, Aquinas didn't think that existence is a perfection. He objects, and Johannes Caterus put the point to Descartes, that the ontological argument doesn't demonstrate that God really exists. It only shows that the concept of existence is inseparable from the concept of God. Descartes' argument is only convincing for the claim that if God exists, God exists necessarily.

Descartes accepts that what he has shown is that necessary existence is part of the concept of God. But this, he responds, is enough: necessary existence entails actual existence. That God must exist to be God means that God exists. But Caterus says this isn't enough: the question is whether the concept of necessary existence entails actual existence.

Kant

Everyone now agrees the problem lies with Descartes' premise (3). What is a ‘perfection'? It's a property that it is better to have than not have. So is existence this kind of property? Descartes and Anselm are supposing that it is - that something that ‘has' existence is greater than something that doesn't.

Immanuel Kant developed the objection to this claim (Critique of Pure Reason, Book II, Ch. 3, § 4). Things don't ‘have' existence in the same way that they ‘have' other properties. Consider whether ‘God exists' is an analytic or synthetic judgment. According to Descartes, it must be analytic: his argument is that ‘God does not exist' is a contradiction in terms, for the concept ‘God' contains the idea of existence (necessary existence belongs to God's essence). But, Kant claims, this is a mistake. Existence does not add anything to, or define, a concept itself; to say something exists is to say that some object corresponds to the concept. To say something exists is always a synthetic judgment, not an analytic one.

When we list the essential properties of something, we describe our concept of that thing. For instance, a dog is a mammal. But now if I tell you that the dog asleep in the corner is a mammal and it exists, I seem to have said two very different sorts of things. To say that it exists is only to say that there is something real that corresponds to the concept ‘dog'. It is not to say anything about the dog as a dog.

Existence, Kant argues, is not part of any concept, even in the case of God. To say that ‘God exists' is quite different from saying that ‘God is omnipotent'. So it is not true to say that ‘God exists' must be true.

necessary existence

If existence isn't a property that something ‘has', then it can't be a property that God has necessarily! And yet it seems plausible to think that if God exists, God exists necessarily. God cannot be a contingent being. If God's existence were not necessary, God would depend on something else that could cause God to come into or go out of existence. If Kant were right, then not only can existence not be a property, necessary existence - as a type of existence - can't be a property. So God can't exist necessarily, even if God exists.

In fact, this doesn't follow. There is still a sense in which God can exist necessarily, if God exists. Rather than saying ‘God has necessary existence', which suggests existence is a property, we should say that ‘it is necessarily true that God exists'. The ‘necessity' applies to the claim: ‘God exists' must be true. Of course, we need an argument to support the claim, but at least it makes sense.

The ontological argument seems to say that because, according to the concept of God, God exists ‘necessarily', that is not contingently, without dependence on anything else, then ‘God exists' must be true. But this doesn't follow; it confuses two meanings of ‘necessarily'.

and morality

Some philosophers believe that morality is objective, and that this cannot be explained unless God exists. For example, in The Moral Argument for Christian Theism, H P Owen argues that the only way to explain how values can exist independently of us is if morality is about our relationship to God.

First, moral demands seem impersonal, not deriving from any particular person or other, yet the ideas of obligation, duty, etc. suggest a personal constraint, while the idea of a moral command or law implies a commander or law giver. Who do we owe moral duties to? Why would we feel guilty or have a sense of responsibility about what we do unless we are responding to someone? Any attempt to explain these apparently ‘personal' aspects of morality naturalistically, e.g. in terms of society, fails to account for the authority of morality. However, an explanation of morality in terms of God secures both authority and the personal nature of morality.

Kant on morality:

Kant presented quite a different argument from morality to God. He rejected the idea that morality is founded on God or God's commands, instead claiming that morality is founded upon reason. (In Owen's terms, Kant is good, therefore, at explaining the impersonal authority of morality, but not its personal nature.) However, Kant argued that we need the idea of God to make sense of our moral duty.

Our duty is to seek the highest good, i.e. one in which moral laws are obeyed and people receive what they morally deserve. Our happiness is not the be all and end all of morality, but it is important. In the best possible situation, acting morally would make us happy; acting immorally would not. Obviously, that situation doesn't exist; but could it possibly? The only way to make sense of the possibility of the highest good is to suppose there exists a supreme being capable of bringing nature (natural events, human happiness) into line with morality. If that alignment doesn't happen in this life, then we may suppose it happens in the next.

Kant does not intend this to be a proof of the existence of God. We should act morally whether or not the highest good is possible; but it becomes questionable how much sense it makes to strive after a goal that we do not think possible. So we need the idea of God to make sense of our practical lives. Since morality is founded on rationality, and God is necessary to make full sense (subjectively, practically) of morality, it is at least rational to believe in God.

Kant's first premise, however, is questionable: is it our duty to seek the highest good? If this is impossible, because God does not exist, then it would indeed seem irrational to pursue it. But this raises no difficulty for duty if we have no duty to pursue it. Perhaps our duty is to pursue as great a good as we can achieve, i.e. to be as morally good as we can be, and seek to do what we can to align human happiness with goodness. A popular principle in ethical theory is ‘ought implies can': you can have no duty to do what you cannot do. By supposing that our duty is to seek the highest good, Kant puts himself in the position of needing to show that we can make sense of such a goal. But if we do not have that duty, we have no need to suppose that God must exist in order to make our goal possible.

However, if we can provide an argument for thinking it is our duty to seek the highest good (perhaps that all our efforts in this direction do logically aim at this end), then Kant's argument carries more weight. But we may raise a second objection: whether rationality really requires us to suppose the goal at which we aim to be achievable. It may be perfectly rational to seek an impossible goal, e.g. science seeks empirical truth - must we therefore suppose that it is possible for us to know everything? This is, at least, contentious.

The argument from religious experience

The argument

Many people have experiences they identify as ‘religious'. Experiences that are part of a religious life include the ups and downs of faith, doubt, sacrifice, and achievement. We are interested in only those experiences in which it seems to the person as though they are directly aware of God or God's action.

Some philosophers have argued that these experiences are importantly similar to perception, an immediate awareness of something other than oneself. We usually treat perceptual experiences as veridical, unless we have good reason to doubt them. Furthermore, the fact that other people have similar perceptual experiences supports the claim that perceptual experiences show the world accurately. Some philosophers have argued that religious experiences are also similar to each other, despite occurring to very different people in very different circumstances. The best explanation of these experiences, and their common nature, is that they are veridical, i.e. they are experiences of something divine. Therefore, God exists.

There are three important questions to discuss. First, what is the similarity between religious experiences, and how do their characteristics support the existence of God? Second, what philosophical problems are there for thinking that these experiences can give us knowledge of God? Third, is there an alternative explanation for the experiences?

James: What is religious experience?

In The Varieties of Religious Experience, William James argued that, for all the apparent differences between religions and religious experience, it was possible to detect a ‘common core' to all (genuine) religious experiences.

1. Religious experiences are experiential, like perception. They are quite different from thinking about God or trying to imagine God's nature.
2. However, they aren't connected to any particular mode of sense perception (sight, hearing, etc.). Sometimes they can be, e.g. the person may feel God is speaking to them; but the ‘inner words' are not normally everything about the experience. They are part of an awareness that transcends sense perceptions, that doesn't have sensory content.
3. The person feels they are immediately aware of and connected to God. This awareness tends to block out everything else temporarily, perhaps even to the degree that the distinction between the person and what they are aware of disappears (‘mystical union').

The heart of religious experience, James argues, is a immediate sense of the reality of the ‘unseen'. By this, he means to contrast what we are aware of in a religious experience with the usual ‘visible world'. Our awareness of the ‘unseen' may be inarticulate, beyond even an ability to think in any usual terms about it. Conceptualization, an attempt to describe it, say wha