The Existence of God

Glenn Mason-Riseborough (Started 26/08/1996; Last Modified 14/08/2003)

 

(This is merely intended to be a collection of notes.  That is, it is not my intention to give a complete or exhaustive account of all possible arguments, and for the most part I haven’t even bothered to give complete references.  It is simply a place where I have striven to give, in broad brushstrokes, the way I see the theism/atheism dialectic proceeding, and my thoughts on how things turn out.  Also, as a collection of notes, this is merely the current version of a continually evolving work.)

 

1.0 General Background

My aim here is to look at and evaluate some of the main arguments for and against theistic belief.  On the “for” side, I am looking at reasons that the theist might give as to why it is reasonable for one to accept a broadly theistic worldview.  On the “against” side, I am looking at arguments that the atheist might give to show that theistic belief is not rational, and hence why one ought not be a theist.

Firstly, when I talk of theism at this most general level, I am primarily considering the three main theistic religions - Judaism, Christianity and Islam, although I would also be open to the possibility of including other religious traditions that could be argued as being broadly theistic (e.g. Zoroastrianism).  Also, when I talk of theism, I am referring to what has been variously called a traditional, classical, or non-radical theistic worldview.  This is the worldview that talks, in a literal and realist sense, about a supernatural realm and a unique god who holds omni-properties (i.e., omnipotence, omniscience, and omnibenevolence), is a person in some meaningful sense (has beliefs, desires, hopes, fears, emotions, intentions, and so on, and is an entity that one might have a personal relationship with), and is creator ex nihilo and sustainer of the universe.  That is, I am working within the assumption that theism entails a belief in what is called omniGod, or the god of the philosophers.  And henceforth, when I use the word “God” in this work I am meaning it as a shortcut to refer to such a being as described.  In this context I do not wish to make heavy weather over this assumption, although in discussions in other places I look at the way that the term “theism” has been fleshed out, and I show that, if we look at its historical usage, it need not have such a restrictive technical meaning.  In addition, I wish to make it clear that for my purposes I am not overly concerned about which theistic religion (or denomination/sect/cult of a particular religion) we should accept (amongst the traditional theisms), and I am prepared to accept philosophical inspiration from any relevant source.  That is, while some details may differ amongst the various traditional theistic denominations, the essence, as belief in omniGod, is constant, and all theisms potentially have insightful things to say on this topic.

Secondly, I wish to clarify what, in my mouth, are meant by words such as “agnostic,” “atheist” and “theist.”  An agnostic, to my way of thinking, is a person who is in a position of not knowing whether or not God exists.  In addition, an agnostic may be in a position of not even knowing whether or not GOD is a coherent concept (though some may claim that it is).  Some agnostics (who we could label sympathetic unbelievers) may desire to be theists if only they could be supplied with reasons in favour of theism.  Other agnostics (more militant unbelievers) may conversely desire to be atheists, but likewise need positive convincing in that direction.  That is, agnosticism itself is neutral with respect to which way one’s personal feelings may lean.  And while agnosticism is primarily about not knowing, it is not not knowing in the sense of ignorance, but rather not knowing in the sense of lacking proof or evidence.  Thus typically one is an agnostic if, after considering the issues, one has found the results inconclusive.  Agnosticism may also be divided into necessary agnosticism and contingent agnosticism.  The former label refers to the doctrine that it is impossible to know one way or the other regarding the existence of God (e.g. Kant); the latter refers to the more conservative claim that it is unknown if God exists or not, but maybe it is possible to find out, somehow.  In other words, agnosticism is generally considered to be a claim about knowledge, or lack of it.  In contrast, atheism and theism are generally considered to be claims about belief, and hence are of a different category of propositional attitudes.  With this in mind, it doesn’t seem incoherent for one to be both an atheist and an agnostic (believe against but don’t know), or for one to be both a theist and an agnostic (believe for but don’t know).

The atheist, in contrast to the agnostic, believes that God does not exist.  Atheism is not to be identified with mere unbelief or disbelief in the existence of God—the state that one is in if one has not heard of God, or has not considered the issue, or has withdrawn from a theistic tradition without reflection, or belongs unreflectively to a non-theistic tradition.  Rather it is a more active denial of the central claims of theism.  Such atheists may hold their belief dogmatically; others may purport to have reasons for their denial of the existence of God.  Those who purport to have reasons may either claim that the theistic concept of God is incoherent, or alternatively, that the instantiation of the concept is unlikely given current empirical evidence.  Many atheists also hold the further belief that theistic belief is unreasonable.  However this is not a necessary characteristic of atheism, and it is possible for an atheist to consistently also believe that it is rational for some people, in some situations, to be theists (but such an atheist happens not to be one of these people).

The theist, in contrast to the agnostic and atheist, believes that God exists.  Like the atheist, a theist may hold his/her belief dogmatically, or may purport to have reasons for his/her belief in the existence of God.  And again, presumably under this definition those who turn up to church each Sunday merely because “it is what one does” and have no beliefs on the matter would not be theists.  As I say, my focus is to identify a number of arguments given by both atheists and theists, evaluating both the theists’ and atheists’ claims.

Thirdly, I should explicitly say that the assumption here is that it is possible to discuss these issues rationally.  As I suggested above, those more inclined towards necessary agnosticism may think all such arguments fail, and they would claim to have a general argument as to why this is so.  As interesting as such arguments may be, I shall not pursue this line of thinking any further here.  Equally, we may get theists who discard all such apologetic arguments in favour of a belief without reason.  These are the type of people who think that faith and reason are mutually exclusive, and further that faith is the virtue and reason is the vice.  These are the type of people who might take inspiration from Tertullian’s claim that the Resurrection of Christ is impossible, and that is precisely why one ought to be a Christian.  Again, I shall not pursue this line of thinking any further.  Those who find this view persuasive will never find dialogue useful, and are unlikely to find these notes beneficial.

Finally, I wish to acknowledge here that I have shamelessly pilfered numerous ideas from John Bishop, in terms of what he has said in private conversation, in lectures he has given at Auckland University, and in a number of his published works.  My ideas are, to a large extent, a reflection of his ideas, though of course I do not claim to represent his views entirely.

 

1.1 The Theist’s Arguments

Over the centuries there has been a plethora of arguments given which claim to prove that one ought to believe that God exists.  While they have come in many different styles and have emphasised many different facets, it seems to me that they can be broken down into a relatively compact set of categories.  Below is what I see as the clearest attempt at providing a taxonomy of apologetics.

 

(a)    Evidential or epistemic reasons, based on claims of truths about the world.  Group (a) can be further divided:

(i)            Appeals to authority, such as appealing to specific scriptures or individuals

(ii)          Appeals to morality, such as claims that we cannot have moral laws without a law-giver

(iii)         Appeals to a priori truths—truths about the way the world must logically be, without looking at empirical facts about the world.  For example the Ontological Argument

(iv)        Appeals based on empirical or experiential facts about the world.  Group (iv) can be further divided:

(1)         Individual experiences, perhaps such as mystical insights or enlightenment

(2)         Appeals to social norms

(3)         Appeals to extra-ordinary events, i.e. miracles

(4)         Appeals to ordinary events.  For example, the Teleological Argument, Cosmological Argument, or Argument from Consciousness

(b)   Prudential or pragmatic reasons, which appeal to self-interest, e.g. Pascal's Wager

 

1.2 The Atheist’s Arguments

The arguments are not just in the direction of the theist trying to convince the sympathetic unbeliever.  There are also arguments given by the decidedly militant unbeliever against theism, and there are various versions.  The Argument from Evil is the main atheistic argument given.  There are two main versions of this—either the attempt to show that traditional theism is incoherent in the sense that one cannot coherently hold all theistic claims at the same time (they are contradictory), or that it is implausible to hold these claims given current empirical evidence (for example certain theistic claims contradict certain claims given by the empirical sciences).  I shall examine the Argument from Evil below in Section 8.

Secondly, there are a number of arguments which have the form “there is no evidence that God exists (that is, all theistic arguments fail), we ought not believe anything on insufficient evidence, therefore we ought not believe that God exists.”  In short, an Evidentialist argument.  I will look more at this form of argument in Section 9.

Thirdly, there are other, more specific, arguments, which attack one or more of the varieties of theism (but not all).  For example, there are a number of arguments aimed specifically against the version of Christianity that accepts Biblical inerrancy.  These are typically of the form “the Bible says both A and not A (in different parts), this is a logical contradiction, therefore at least some of the Bible must be false and cannot be inerrant.”  Given that I am focusing on theism as a whole, I shall not pursue these more specific arguments any further.  Also, given that the assertions about the Bible, for example, are not theistic assertions (as I am using the word here), these arguments are not atheistic arguments.  Perhaps they are best understood as anti-certain-varieties-of-Christianity arguments, for example (and there are corresponding arguments against Judaism and Islam), and thus they may potentially hold weight against certain alternative varieties of religious belief.

Fourthly, some arguments (which are very close in structure to the Argument from Evil) don’t so much argue that God does not exist, but argue that we ought not worship God or follow his instructions.  These typically take the form “God (assuming he exists) has done a (where a might be specified either in terms of clearly observable natural phenomena, or in terms of certain revelations), a is morally wrong (specified either as something we would all intuitively agree upon, or as argued from some more general moral theory), therefore God is not morally perfect, and ought not be worshipped.”  These, strictly speaking, are neither arguments against believing that God exists, nor arguments against worshiping omniGod.  Thus, strictly speaking, these are not atheistic arguments, and I shall not pursue them further here.  However, like the third type of atheistic argument, they may still hold weight against certain alternative varieties of religious belief.

 

1.3 My Claim

I should say at this point that I don't think that any of these various theistic or atheistic apologetic arguments work.  That is, I don't think that any of the theist’s arguments are good enough to convince the sympathetic unbeliever, and I don’t think the atheist’s arguments are good enough to show that theistic belief (as a whole) is irrational.  As I see it, at best we get to an impasse between different worldviews (naturalism vs. traditional theism) – the theist is unable to show that traditional theism wins out over naturalism, and the atheist is unable to show that naturalism wins out over traditional theism.  This places me squarely in the agnostics’ camp (as a contingent agnostic), in the sense that I don’t know that God does or doesn’t exist.  However, I must admit that, firstly on the basis of Ockham’s Razor (the principle that broadly states we ought not explain something by appealing to more entities if a simpler explanation with fewer proposed entities has similar explanatory power) I find naturalism more plausible, and secondly on the basis of ethical arguments, I find omniGod theism implausible (though not impossible).  So, practically speaking, I live with the assumption of God’s non-existence, and probabilistically speaking, I consider myself an omniGod atheist (I believe that God probably doesn’t exist).  But I accept that there is some ground for theists to accept theistic premises that I personally find implausible, and I see insufficient grounds to go the further step and claim that it is irrational to be a theist.

However, I don't want to be unduly dismissive of these theistic and atheistic arguments.  It seems to me that the arguments should not be seen merely as arguments for theism or atheism, but also as useful (partial) descriptions of the respective worldviews.  That is, as arguments they are fundamentally flawed because, at best, they already assume their respective worldviews (they are circular), but as descriptions of those worldviews they are potentially very insightful.  This is because they are able to tell us something of the assumptions inherent in the respective worldviews.  Often the arguments appear so convincing to those already within the respective camp because the believers are already in a position of assuming a hidden (or maybe a not-so-hidden) premise so utterly that they don’t conceive of or allow for the opposite view.  The problem is that those in the opposing camp are equally so convinced of the negation of that assumption (to my mind, this comes out clearest in arguments such as the Argument from Evil and the Cosmological Argument).  It is only by examining the arguments thoroughly that we are able to tease out these assumptions.  Of course, bringing out these assumptions is unlikely to result in agreement between the two camps, but at least we may have been able to narrow down slightly the point of disagreement.

 

1.4 My Plan

My plan is to structure this to vaguely follow my taxonomy of apologetics.  To date, I include in various stages of (in)completeness (maybe one day I’ll finish this):

 

·        Section 1: Introduction

·        Section 2: Authoritative Theology – Authority, Experience and Society [not started]

·        Section 3: Moral Theology – Morality [not started]

·        Section 4: A Priori Theology – Ontological Argument

·        Section 5: Supernatural Theology – Miracles

·        Section 6: Natural Theology – Teleological, Cosmological, and Consciousness Arguments [sketchy and dated]

·        Section 7: Prudential Theology – Pascal’s Wager

·        Section 8: Natural Atheology – The Argument from Evil

·        Section 9: Conclusions (Existential Theology/Atheology?) [not started]

 

2.0 Authority, Experience, and Society

 

 

3.0 Morality

 

 

4.0 The Ontological Argument

Let’s step back and look at the big picture.  We are looking for reasons that the theist might give to the sympathetic unbeliever as to why it is reasonable to believe that God exists.  The Ontological Argument is an attempt at an a priori argument.  That is, the claim is that if we have a proper understanding of what we mean by God, then we must recognise that God necessarily exists.  According to the theist who puts forward the Ontological Argument, the atheist’s mistake is that s/he misunderstands the nature of God.  There are a number of different versions of the Ontological Argument, historically the two most noteworthy are by Anselm (Proslogian, Ch. 2 – 4) and Rene Descartes (Fifth Meditation).  More recently Alvin Plantinga and others have attempted modal versions, which J. L. Mackie strongly critiques in The Miracle of Theism.  I will focus primarily on Anselm’s version.

 

4.1 Anselm’s Version

Anselm sets out his proof in the form of a prayer.  In his prayer, he says that he has faith, but he also asks God to give him evidence that He exists, such that when the fool (atheist) sees this proof he can likewise believe.  He concludes by thanking God for showing him the proof.  As such, even though the text is structured as a prayer, it is clearly intended as an apologetic argument.  Anselm starts out with the assumption that what we mean by God is that He is the greatest conceivable being, and since a being that exists in reality is greater than one that only exists in our minds, God must exist in reality.  The usual interpretation of Anselm’s version of the Ontological Argument is as a reductio ad absurdum. The formal argument is (taken from Plantinga, 1967):

 

1.      God exists in the understanding, but not in reality [reductio]

2.      Existence in reality is greater than existence in the understanding alone [premise]

3.      A being having all of God’s properties plus existing in reality can be conceived [premise]

4.      A being having all of God’s properties plus existing in reality is greater than God [from 1, 2]

5.      A being greater than God can be conceived [from 3, 4]

6.      It is false that a being greater than God can be conceived [definition of ‘God’]

7.      Hence it is false that God exists in the understanding but not in reality [by reductio]

 

This argument has puzzled many people.  At first glance it seems so convincing, yet at the same time many (including theists) have been reluctant to accept an argument that seems to be too easy.  Surely proving God’s existence is not so simple that it can be done merely through examining our definitions.  Yet, picking out precisely what is wrong with the argument turns out to be fairly difficult, and centres around some very subtle points.

Very soon after Anselm produced his proof, a fellow monk, Gaunilo, replied with a set of criticisms.  The clearest of these criticisms was in the form of a counter-example, which, while not identifying where Anselm went wrong, tried to point out that Anselm could not be correct.  Guanilo imagined a ‘lost island,’ which was superior to all other islands.  Guanilo claimed that if Anselm’s proof of God’s existence was correct, we could likewise claim that this lost island exists.  The idea was that the greatest lost island would be one that exists rather than doesn’t exist, so this lost island must really exist.  But surely this is ridiculous, so, likewise Anselm’s proof of God must be wrong.  Anselm replied by maintaining that there was a real difference between God and the lost island.  A being than which nothing greater cannot be conceived cannot be conceived not to exist since in that case it would be less great than a being that was conceived.  In the case of the island, it was possible that it could be conceived of not existing, but in the case of God, He could not be.  In the case of God, the greatness is qua being, and thus we cannot conceive of the greatest conceivable being not existing, but in the case of the island the greatness is qua island, and we can conceive of the greatest conceivable island not existing.  So, Guanilo’s reply does not work.

Another reply that has been put forward centres around whether it is possible to conceive of God at all.  If it is not possible, then He would not even exist in the understanding.  This is a difficult issue to decide on conclusively either way, so I’ll move on to what I think is the best reply, the Kantian objection.

 

4.2 Kant’s Objection

Immanuel Kant’s objection, while specifically a reply to Descartes’ version, also applies to Anselm’s.  Kant’s reply is in the Critique of Pure Reason, Second Division, Book II, Chapter III, Section 4.

The central slogan which is associated with Kant’s objection is that “being [or exists] is … not a real predicate.”  The thought here is that while ‘exists’ is clearly a grammatical predicate, it is not a real, or determining predicate.  That is, we don’t add anything to the concept of a thing – we don’t determine the concept any further, when we add the description that it exists.  Kant uses the example of a hundred thalers.  He points out that we don’t tell ourselves anything more about the concept of a hundred thalers if we then add that it exists.  More succinctly, he says, “a hundred real thalers do not contain the least coin more than a hundred possible thalers.”  So, according to Kant, what we are saying when we claim that something exists is that some determinate concept is instantiated.  For example, we have something that has the properties F, G and H.  When we assert that such a thing exists, we don’t add a further property to the list determining what the thing is (i.e., we don’t say F, G, H, and E).  Rather we are concerned that there is something that has the properties F, G and H.  In symbolic notation we use the existential quantifier: .

So, Kant claims that the Ontological Argument is false because it mistakes superficial grammatical structure for deep logical structure.  Kant claims that ‘exists’ cannot be used as a real predicate, it can only be used to express the instantiation of the concept.  But the Ontological Argument proceeds as if ‘exists’ were a real predicate.  But does Kant’s objection work?  Surely it seems very dogmatic!  Surely it is the case that sometimes ‘exists’ does refer to a property that a thing does or does not have.  The example that we might use is when we compare real characters with fictional characters.  The difference between an actual and a fictional character is that the actual character exists and the fictional one does not exist.  For example, we might say that Queen Elizabeth II exists, whereas Santa Claus does not, and we are clearly adding something to the meaning when make this further claim.  Thus ‘exists’ can be used to determine a concept.  Mackie shows this distinction clearly by the use of his Remartian example.  A Martian is an inhabitant of Mars, whereas a Remartian is an actual existing inhabitant of Mars.  We further determine the concept of a Martian when we add to it to in order to produce the concept of a Remartian.  We can tell a fictional story about Martians, but we cannot tell a fictional story about Remartians.

Put this way, Kant’s critique has not succeeded and we must upgrade his claim to take this into account.  We have to modify Kant’s claim that ‘exists’ is not a real predicate.  It is not that ‘exists’ is never a real predicate, but rather ‘exists’ has two uses.  Firstly, the descriptive (real/determining) use, which refers to a highly general and abstract property, and secondly the instantiatory or existential use.  So, the corrected slogan should be: ‘Exists’ is not only a real predicate, it is the instantiatory ‘predicate,’ or existential quantifier, and, with respect to every concept C, it always has the function as the instantiatory predicate, or existential quantifier, because the question can always be raised whether there exists anything which is C.

This shows us that the Ontological Argument is not a good argument in that it is unable to prove that God is instantiated in this world.  But, this modification of Kant’s critique allows us to concede that Anselm’s Ontological Argument does tell us something very important.

 

4.3 Points To Learn from the Ontological Argument

The Ontological Argument explores the concept of God as found in the religious context itself (recall: for Anselm the Ontological Argument occurs in prayer).  And its exploration yields two claims regarding our conceptual understanding of God.

Firstly, Mackie observes that if we understand God as the greatest conceivable being, then God’s essence includes God’s existence.  While our modified Kantian critique showed us that the Ontological Argument did not work as a proof of God’s instantiation, it showed us that part of what we mean by God is that He cannot be a fictional entity.  That is, it didn’t show us that God exists, but it did show us that if anything is God, then it exists.  To contrast, we can completely understand the concepts of unicorns, quarks, and triangles, while recognising that unicorns, quarks, and triangles could turn out to be purely fictional entities.  We cannot do this with Remartians or God.  But to further contrast, in the case of Remartians, existence was simply tacked on, but in the case of God existence is an integral component of the description of God’s nature.

Thus, we might use the Ontological Argument against non-realists such as Don Cupitt and Lloyd Geering.  Most typically, the non-realist claim is that talk about God is a useful fiction.  God doesn’t really exist, but it is useful to act as if He did, in order to adhere to certain ethical or cultural values.  So, we might say that if we take God as being the greatest conceivable being, then either the non-realist misunderstands what we mean by God, or s/he has a radically different concept of God in mind.  That is, either s/he misunderstands what it means to be the greatest conceivable being, or s/he rejects the claim that God is the greatest conceivable being.  So, if we accept Anselm’s definition of God, then to count as God, something must at least be actual.  Worship demands as object a reality outside of the self.  As a further issue, each theistic tradition would need to be examined as to whether or not a non-realist concept of God can rightly be considered to be within that tradition.  That is, if we accept Cupitt’s or Geering’s claims, then we would have to show that Anselm is incorrect to insist that the Christian god is defined as that which no greater can be conceived.

The secondly issue that the Ontological Argument brings up is whether God’s essence includes God’s necessary existence.  The claim here is that one hasn’t really understood the concept of God if one thinks that God might turn out to be a purely contingent entity.  That is, to count as God, something must belong to all possible worlds.  I don’t want to discuss this second claim in any depth except to say that while it may be correct, it is questionable.  It will only be correct if a necessary being is greater qua being than a contingent being, and it seems to me that it is a contestable matter that to be the greatest conceivable being, God must be necessary as well as actual.  One possible motivation for rejecting this second claim is if we take seriously the idea that love is the greatest.  Of course, love between two people is a contingent matter since they may never have met or even existed, and so if love really is the greatest being, then necessity is not a great-making property, qua being.  This leads us into a possible alternative concept of God in which God is literally seen as love (see John Bishop, for example).

 

4.4 Conclusion

So, the Ontological Argument is an attempt to show that God must exist based on the definition of God alone.  The claim is that atheists misunderstand what ‘God’ means.  The modified Kantian critique is a successful reply to this claim, stating that no matter how much we add to the definition of something, it is a separate matter as to whether it is instantiated or not.  All the Ontological Argument shows is that if anything is God, then it exists –  and arguably, if anything is God, then it necessarily exists – .  But it does not settle the question whether there is something that is God – .  So we can endorse a revised Kantian critique.  For all concepts, F, the question of whether  is true cannot be settled by analysis of the concept F, even when this analysis shows that E is included in the essence of F (i.e., ).  Thus, we may concede that the Ontological Argument establishes  (and arguably ), yet reject it as a proof of .  So, the atheist and the realist theist do agree on the concept of God.  Anti-realist theists don’t agree, and, at best, have a different concept of God.

 

5.0 The Argument From Miracles

What we are interested in is the reasons that a theist might give to the sympathetic unbeliever as to why he is a theist.  One reason that might be given is the existence of miracles.  The theist might claim that it is reasonable to believe because we have evidence of certain events in this world that cannot be causally explained other than by appealing to supernatural forces.  My aim here is to briefly consider this line of thinking and to consider whether it is ever justified to have a theistic belief on the basis of testimonies of miracles.  In particular, I wish to examine David Hume’s critique of this argument, which he delivers in Section 10 of An Enquiry Concerning Human Understanding, which is entitled “Of Miracles.”  Hume’s claim is that it is never justified to believe in a religion on the basis of reports of miracles, and he prides himself on having a decisive argument against all claims of miracles.  Broadly speaking, Hume’s method is twofold.  Firstly, Hume sets out to show that it is always possible to question the truth of the testimony of the miracle.  Secondly, Hume suggests that even if we should accept the truth of the testimony we would first want to try to explain it in naturalistic terms.  From the theist’s perspective, not only must the truth of the testimony be assured, but also it must be shown that it can only ever be explained by appealing to the supernatural.

To start with, I need to define what we mean by “miracle.”  The first and original sense of the term simply refers to something surprising or marvellous.  The second sense of the term refers to a violation of the laws of nature.  For example, I could look out of my window at the beautiful clear blue day and say that it is a miracle that the world exists and that I am alive.  This is a miracle in the first sense, because, I could nevertheless give a plausible explanatory story in purely naturalistic/scientific terms that says how I came to be here, and my present psychological state.  On the other hand, if I produce something out of nothing, make a dead person come to life, or, in short, do something that goes against the laws of nature, then this is a miracle in the second sense.  For the purposes of using miracles as a reason for theistic belief it is clear that only the second sense will do.

I think it is important to explicitly point out that Hume is not attempting to give an a priori argument that miracles don’t exist (as some have claimed).  He is not trying to show that the notion of a miracle is incoherent, but rather he is attempting to give epistemic reasons to show that we never have sufficient reason for believing that they have happened.  Thus, we need to be careful about how we talk about a law of nature.  We don’t want to say that a law of nature is merely a regularity, or a statement about how the world regularly works.  If this were so, if some regularity was broken, then it would not really have been a law of nature, and the supposed miracle would not really have been a miracle.  But, what we are wanting to do here is contrast the order of nature with a possible divine or supernatural intervention.  So, the laws of nature describe the state in which the world works when left to itself.  Thus, a miracle is an event in which some supernatural force has intervened in the natural ordering of the world.

 

5.1 Hume’s First Argument

Hume's main argument regarding whether to believe reports of miracles is based on weighing up the probabilities given the evidence we have before us.  Supposing someone testifies that a miracle has occurred. On the one hand we have got the unlikelihood of the testimony being false and on the other hand we have got the unlikelihood of the miracle occurring.  There are three possibilities.

 

(1)     The unlikelihood of the testimony being false is greater than the unlikelihood of the miracle occurring.  We must therefore accept the truth of the testimony with a degree of confidence that corresponds to the difference between the two probabilities.

(2)     The unlikelihood of the testimony being false is equal to the unlikelihood of the miracle occurring.  In this case we must suspend our judgement.

(3)     The unlikelihood of the testimony being false is less than the unlikelihood of the miracle occurring.  We must therefore reject the miracle report with a degree of confidence that corresponds to the difference between the two probabilities.

 

Hume's claim is that a miracle, because it contradicts the conclusions of an inductive argument, must be as unlikely as anything could be.  So, if we are to believe a testimony regarding a miracle then there is a lot of work to do to show that the unlikelihood of the testimony being false is greater.  Hume states: “no testimony is sufficient to establish a miracle, unless the testimony be of such a kind, that its falsehood be more miraculous, than the fact, which it endeavours to establish.”

Hume's next move, which he carries out throughout most of Part II of the section, is to show in various ways how a testimony may be false.  His claim is that because of these reasons, in practice, no testimony is ever strong enough to warrant either options (1) or (2) above.  Firstly, Hume states that no one is ever completely trustworthy, i.e. no one ever has such a strong reputation that s/he is beyond suspicion with regards to deceiving others.  Secondly, Hume claims that our natures are such that we have a liking for the miraculous – “the passion of surprise and wonder, arising from miracles, being an agreeable emotion, gives a sensible tendency towards the belief of those events, from which it is derived.” Thirdly, belief in miracles most often occurs in “ignorant and barbarous nations.”  Where these beliefs occur in civilised people, it is because it has been passed down from ignorant ancestors.  Fourthly, different religions conflict with each other, effectively cancelling each other out.  Fifthly, miracles are often used to introduce a new religion or provide support for an existing one.  Experience shows that many people are deluded by such events.  In fact, it is often the case that religions promote credulity as a virtue and frown on critical caution as a sin.  It is clear that some of these arguments are weaker than others, yet Hume seems to give some useful psychological reasons for putting at least some doubt on testimonies.  It is not my purpose in this paper to examine each of these arguments in detail, but suffices to say, I think that taken together Hume’s reasons build to strongly support his claims.

 

5.2 Hume’s Second Argument

At the end of Part II, Hume briefly discusses another argument.  In this second argument Hume accepts the possibility that maybe there are times when the testimony is so secure that we have to accept option (1) above.  The main part of this argument consists of a scenario.  In this scenario Hume supposes that from the first of January 1600, for a period of eight days, there was total darkness over the entire earth.  We are assured of the truth of this because we have historical records from numerous independent sources from all over the world, in many different languages.  Hume suggests that in this case the evidence of testimony would be so strong that the unlikelihood of the testimony being false would be greater than the unlikelihood of the event happening.  Hume's claim here is that even if we should be faced with a prima facie inexplicable event – a “miracle,” rather than claiming that the event had supernatural origins, we would instead set about looking for a naturalistic explanation of the events (“search for the causes from whence it might be derived”).  Hume doesn’t say much more about this argument, but I think it is worth making a few more comments.  It seems to me that this approach that Hume suggests is precisely what has happened in the history of science to date.  At various times people have discovered anomalies between observed events and predicted events (based on the then current understanding of the laws of nature), and as a consequence have had to modify their theories to encompass the observations.  To date we have been extremely successful in doing so, and our body of scientific knowledge has grown considerably.  Yet, it seems to me that even if we find that we are unable to come up with an adequate naturalistic explanation, we might still wish to claim that there was some naturalistic explanation.  In fact, even if we were to say that due to our limited human capacities it is impossible to ever find a complete naturalistic explanation of some event, this does not mean that Hume’s argument has failed.  Why do I say this?  Well, as I see it, what is at issue here is two completely different worldviews.  On the one hand we have the purely naturalistic worldview, and on the other we have the worldview that also includes the occasional (or not so occasional) intervention by the deity.  We have something of an impasse between these two views.  We cannot say that miracles are impossible, but neither can we say that they happen.  What we can say though is that this impasse provides no hope for the theist who is wishing to convince the sympathetic unbeliever on the basis of miracles.  In reply to the theist's claim towards the supernatural, the sympathetic unbeliever will point out that there are possible naturalistic explanations.  The theist is unconvincing because s/he is unable to claim that appealing to the supernatural is the only possibility.

 

5.3 Conclusion

It seems to me then that Hume has produced an extremely strong argument.  He is requiring the theistic apologist, who is claiming belief on the basis of miracles, to defend two weak positions.  Firstly, the theist has to show that the testimony of the miracle is so strong that the unlikelihood of it being false is greater than the unlikelihood of the miracle occurring.  Secondly, Hume is requiring that even if the theist is able to show that the testimony is indubitable, the theist must then also show that appealing to supernatural intervention is the only possible explanation of the event.  Thus, on the whole, I think that Hume’s argument is successful.  I think he shows that the appeal to miracles as the reason for theistic belief is not rationally justified.

 

6.0 Natural Theology

 

 

6.1 The Argument From Consciousness

 

 

6.2 Teleological Argument

One small clarification, the “Argument from Design” is a misnomer.  It is a trivial matter to move from design to designer.  What we have to do first is move from apparent design features to design.  This is the big step.  Thus, is it better called Argument for Design, or Teleological Argument.

In the case of the Teleological Argument there are two main versions - an argument by analogy and an inference to best explanation.  In the case of the former version I think Plantinga correctly dismisses it, primarily because it is impossible to know to which set of objects the universe best belongs.  Why I say this version might be circular is if the theist implicitly thinks that the universe clearly best belongs to the set of designed objects, whereas the naturalist would claim otherwise.  In the case of the latter version, the sympathetic unbeliever would be likely to point out the remarkable advances in the sciences, and that they give a completely coherent explanatory story.  The theist would counter this claim, perhaps by saying that even if we accept the current scientific worldview, it is still possible to postulate a creator who set in motion the big bang (or whatever favourite naturalistic theory the theoretical physicist puts forward).  Both views are possible, and we have an impasse.  Hence the circularity if the theist continues to advocate the theistic worldview, and hence the argument does not work as an apologetic argument.

 

6.3 The Cosmological Argument

This moves us into the Cosmological Argument (if we take the point of design further enough back).  One point of impasse is that the theist will often assume, implicitly, that an infinite chain is incoherent, and thus always seek to find some "first cause."  But as I see it, the main point of impasse is best seen by asking whether it makes more sense to say that the physical order is self-explanatory, or that the mental (God) is self-explanatory.  The naturalist will say that the theist is continuing to implicitly think theistically if he keeps asking the question - "well, what made the natural universe?"  The naturalist sees it as coherent to maintain that the natural universe contains within itself the reason for its existence, and thinks it bizarre to need to ask the further question.

 

6.3.1 Aquinas’ Second Argument

In the third article of his Summa Theologica, Saint Thomas Aquinas presents five arguments (‘ways’) from which he claimed that God must exist.  In this section I shall examine Aquinas’ second argument – the argument from causality.  This argument is very similar in structure to Aquinas’ first argument – the argument from motion, and in many ways it is simply a more general form of it.  My emphasis will be on establishing whether or not the second argument satisfactorily proves what it set out to prove, namely God’s existence.  I will identify weaknesses in the argument, explain why are they weak, and why Aquinas might have thought that they were valid.  Aquinas’ second argument, as given by Elliott Sober (1995, p. 37), is:

 

1.    Some events cause other events.

2.    If an event happens, then it must be caused by something outside itself.

3.    There can be no infinite cause/effect chains.

4.    So, there is a first, uncaused cause.

5.    Therefore, God exists.

 

Starting at the beginning, premise 1 states that some events cause other events.  This seems a straightforward statement based purely on observation.  If we observe some event E₁ happening, we may find an event E₂ such that event E₂ caused event E₁.

Premise 2 then goes on to state that event E₁ cannot equal event E₂.  This means that some event cannot cause itself to occur.  This seems to be intuitively obvious, if I press a key on my computer keyboard, then pressing the key cannot have caused me to press that key.  Premise 2 also states that any event E₁ must have been caused by some event E₂.  Pressing the key was caused by some motor action in my arm (amongst other things), and if I care to name any event, then there must be some other event that caused it.

Unfortunately, once we get to premise 3 we get to the problem of the infinite regress of causal chains.  It states that we cannot have a chain of events such that some event E₁ was caused by event E₂, E₂ was caused by E₃, ..., E_x was caused by E_x+1.  Presumably, Aquinas’ reason behind this is that time cannot go back into the past forever – one would never get to ‘now’ if past time was infinite, and the universe had no beginning.  There are a number of possible replies to this.  Firstly, we might say that even though the cause/effect chain is infinite, this chain exists in finite time (as we go back in time, each event takes less time, much like Zeno’s paradox).  Secondly we might suggest that if time is somehow curved, and curved to such an extent that the chain of cause and effect is an unbroken ring.  The second law of thermodynamics seems to dispute this theory, stating that entropy is forever increasing) (things decline into chaos in a closed environment) therefore a circle of increasing entropy cannot exist.  A fix might be to use more exotic physics such as superstring theory to propose that since higher dimensions exist (other than three space and one time), then the universe as we know it is not a closed system.  A third related way to get around the infinite series problem is to use quantum theory to say that causality is not linear (and in fact this might even persuade us to reject premise 2).  This tells us that some event in the present does not have to be caused by some other event in the past.  At the subatomic level there is evidence to suggest that some particles may interact in such a way that the effect occurred temporally before the cause.  This would mean that although the cause/effect chain is infinite, the universe may have begun some finite period of years ago.  All these theories indicate that there may or may not be an infinite chain of causation, and that causality may be irrelevant when considering the origins of the universe.  Of course Aquinas would not have been aware of these theories and his reasoning would have occurred without this additional knowledge.

From these first three premises, Aquinas made an intermediate conclusion, premise 4.  This premise states that since there can be no infinite cause/effect chain, then at some point there must be some first, uncaused cause, let’s call it event E_y.  So the structure is now E₁ was caused by E₂, E₂ was caused by E₃, ..., E_x was caused by E_x+1, ..., E_y-1 was caused by E_y.  Unfortunately this statement simply contradicts premise 2, which states that there cannot be an uncaused cause (all events need a cause), so E_y cannot exist.  In fairness to Aquinas, premise 2 may be changed to state that there can be no uncaused cause except for the first cause, and this is probably what Aquinas meant.

The final conclusion drawn from premise 4 is premise 5; God exists.  This conclusion simply does not follow.  We cannot conclude from there being an uncaused cause that God exists.  This uncaused cause may be the Big Bang, or it may be Descartes’ evil demon (which does not hold the property of being omnibenevolent).  Adding the premise “God is the first, uncaused cause” (let’s call it premise A) makes the argument deductively valid, and this is what Aquinas must have believed.  However, since there is no evidence for this premise being true, we must say that it is a weak argument.

An argument is only as strong as it’s weakest link, so by looking at the links of this argument we can decide if it is strong.  Premises 1 and 2 seem strong enough, however premise 3 is extremely questionable.  It may or may not be true, we just do not know either way.  Premises 1, 2, and 3 simply do not imply premise 4, and 4 contradicts 2.  Altering premise 2 only results in us doubting its truth.  Premise 4 does not imply premise 5, and adding premise A only results in introducing another questionable premise.  We are therefore forced to say that a combination of false or weak premises and weak links result in an extremely weak argument.  The only reason we have of understanding why the argument may seem convincing is to realise that Aquinas would not have had modern physical theories.   Also he would have held true many of the hidden assumptions that on inspection I have questioned.

 

7.0 Pascal’s Wager

Pascal’s Wager is the name given to an argument developed by Blaise Pascal for believing, or taking steps to believe, in God.  Pascal presents this as one of his arguments in Pensées (there are, in fact, at least three separate arguments in Pensées that might be wagers, generally only the third of which is called Pascal’s Wager).  Though at times Pascal is extremely unclear (Pensées was never finished and was left as notes) and even appears to contradict himself, nonetheless the Wager has taken on a life of its own and it is possible to present it in formal terms.  Pascal makes the radical claim that all arguments that attempt to prove or disprove God’s existence are inconclusive.  Thus, the Wager does not attempt to give evidence to support the claim that God exists.  Pascal’s approach is to instead provide prudential reasons for believing in God.  The Wager is an attempt to show that belief in God is in one’s own best interests.

 

7.1 Formulation

To use modern (Bayesian) terminology, Pascal formulates his Wager in terms of decision theory.  Both the possible relevant states of the world and the possible actions of the agent are considered.  Often this is depicted graphically as a decision matrix, in which the columns represent the former and the rows the latter.  Utilities, numbers that correspond to the degree to which the agent values them, are assigned to each box in the matrix.  In decisions under risk, the columns are additionally given probability weightings; in decisions under uncertainty, no probabilities are given.  When considering decisions under risk, rationality requires that one should choose the row with the greatest expected utility (EU).  That is, one should choose the row with greatest total value when the utilities are multiplied by the probabilities and are added across the rows.  When considering decisions under uncertainty, rationality requires that one chooses the row in which the worst outcome is at least as good as the best outcome of the other row.

I will only consider Pascal’s third wager, which is a decision under risk.  Pascal claims that either God exists or He doesn’t and that we must either choose for God or against God.  No other option is possible.  Thus, the Wager is presented as a 2x2 matrix.  If God exists, one gains all if one wagers for God’s existence but loses all if one wagers against Him.  If God doesn’t exist then at worst the loss or gain will be finite.  The probability of God’s existence is non-zero.  The decision matrix is:

 

	God Exists	God Does Not Exist
Wager For God	Infinite Gain	Finite Loss
Wager Against God	Infinite Loss	Finite Gain

 

7.2 Possible Objections

There are numerous objections that have been raised at different times by different people.  What follows below is simply a listed summary of what I consider to be the main ones.

1.      The Utilities may be questioned:  (a) One may question whether it is possible to have an infinite utility.  Some people have argued that the notion of infinite utility is suspect, either by saying that the utility of salvation must be finite or that human conception is so limited that infinite reward could only be finitely appreciated.  (b) There may be different utilities for different people.  God’s “Chosen People” may be predestined for reward but others may not, no matter what they believe.  Alternatively, salvation might appeal to some people more than it does to others.  (c) We might suggest that we can still have infinite reward on earth, even as finite beings in the face of God’s non-existence.  This is achieved by living our lives in such a way that we are fully flourishing as finite human beings.  Perhaps it could be further said that those who wager for God’s existence are not fully flourishing human beings and have infinite loss.

2.      The matrix should have more columns (Many-Gods objection):  It is most likely that Pascal had a Catholic concept of God in mind, but it seems that the “God Does Not Exist” column could be subdivided into various other theistic hypotheses.  In fact, one might say that there are infinitely many Gods to consider.  We might also need to consider options such as 1(c) above and reincarnation into this finer grained distinction.

3.      The matrix should have more rows:  Is there more than one way to wager for God?  If so, might God’s rewards vary?  For example, He might give less reward to those who wager for the mercenary reasons that Pascal gives.

4.      Assigning a probability for God’s existence:  (a) The probability of God’s existence may remain undefined because it is impossible to assign.  To assign a probability is to feign having evidence that one, in fact, lacks.  (b) Strict atheists might argue that the probability of God’s existence is zero (contrary to Pascal’s claim that there is no evidence).  That is, it might be argued that the theistic concept of God is incoherent.

5.      Rationality does not require maximising expected utility:  (a) There are paradoxes that are said to show that maximising expectation can lead one to perform sub-optimal actions.  Consider the Prisoner’s Dilemma.  (b) We might also distinguish between practical rationality and theoretical rationality.  The first requires one to maximise expected utility, but the second requires proportioning belief to the amount of evidence available.

6.      The argument is invalid:  It has been argued that the Wager is not even valid.  At the most extreme, even if you devote all your energy to avoiding belief in God, there is still a non-zero probability of your efforts failing and, despite your best intentions, wagering for God.  Even in this case there will be an infinite expected utility, and thus there is no single maximum expected utility.

7.      Moral objections:  Even assuming it is prudential to wager for God, it does not follow that one should wager thusly.  It might not be moral to wager for God.  For example, the putative divine plan might itself be morally wrong.  Or perhaps believing something on insufficient grounds harms society by promoting credulity.  Or thirdly, maybe both wagering and appealing to self-interest are unworthy of theistic belief.  To overcome this objection we need to explain why rationality trumps other considerations.

8.      What does it mean to “Wager for God”?  Even if one is convinced by the Wager, it still may be difficult, or even impossible, to become a believer, and rationality cannot require the impossible.  Pascal considers this objection and argues that through acting like a believer one will eventually be a believer.  We might question Pascal’s theory of the psychology of belief formation and suggest that some may fail.

 

7.3 Conclusion

Not all the above objections work, but, to the extent that enough of them do it is generally considered that the Wager does not succeed as an apologetic argument for theism.  One way of summarising plausible parts of the first three objections in the form of the decision matrix is to change the utilities as follows:

 

	God Exists	God Does Not Exist
Wager For God	Possible Infinite Gain	Infinite Loss
Wager Against God	Infinite Loss	Possible Finite Gain

 

The upper-right utility is a summary of the flourishing objection (1c); the lower-right utility is a summary of the many-Gods objection (2); the upper-left utility is a summary of the many-rows objection (3).  When these modifications are made, it is clear that both rows have the same expected utility, and rationality does not require wagering for God.

 

8.0 The Argument from Evil

The Argument from Evil is an argument presented by the atheist against theistic belief in omniGod.  The atheist who presents this argument wants to claim that theistic belief is unwarranted or unreasonable, since the object of belief, God, is (depending on how the argument is formulated) either (a) conceptually incoherent or (b) extremely unlikely to be instantiated in the world.  The Argument from Evil is to be distinguished from the Problem of Evil.  The former is an atheistic argument against belief in the existence of omniGod.  The latter is a theistic paradox or dilemma – how could there be evil given that God exists?

There are various formulations of the argument.  What they have in common is the thought that there is, at the very least, a tension between God being all-powerful (including all-knowing), his being all-good, and the observation that the world contains at least some undesirable states of affairs.  Clearly then, such an argument is pitched against a fairly specific concept of God – a maximally powerful and a maximally good moral agent.  It may also worry other concepts of God to a greater or lesser degree, depending on their commitment to power and goodness.  Whatever the case, it is clear that if one lessens God’s power or his goodness, then the tension is minimised or even drops away completely.  A believer in God can rather simply avoid the argument by biting the bullet with respect to God’s power or goodness.  In other words, the Argument from Evil is only an argument against the rationality of omniGod theism; it is only an argument for omniGod atheism, not atheism outright.

 

An initial formulation

Sometimes the argument is presented as there being a logical contradiction between three separate claims.  These are: (1) God is omnipotent, (2) God is omnibenevolent, and (3) there exists evil.  One cannot hold all three claims at the same time; it is an inconsistent triad.  Formulated in this way, the omniGod theist either has to deny (3), or deny the logical incoherence of holding all three positions.

 

A preliminary theistic reply

Turning first to the possibility of denying (3).  What do we mean when we say that evil exists?  The word “evil” can potentially be a very strong and emotive term, so it is important to be clear about what is and isn’t being claimed.  Some people like to define “evil” as intentionally bad acts performed by morally free beings, or perhaps as going against the wishes of God.  Others widen it to include certain other acts or states of affairs.  Which is the preferable definition, in this context?  What we need to do is look at the intent of those who present the argument, and the intent is quite simple.  It is that we can all see things in this world that we can generally agree would be better if they weren’t there.  There is bad or undesirable happenings in the world, and why didn’t or doesn’t God prevent this bad stuff.  Why didn’t he create a better world, or why didn’t he intervene to stop it before it happened?

Sometimes this bad stuff is intentionally bad acts performed by morally free beings, and sometimes it is not.  Occasionally we accidentally and unintentionally do something that we would rather not have done.  Or sometimes unpleasant natural events happen in the world that appear not to have been caused by us humans.  One plausible definition that satisfies the intent of the argument is to say that evil is serious suffering.  It is thus possible to distinguish between two main types of evils.  Moral evil is the serious suffering caused by the actions of morally free agents.  Natural evil is serious suffering that is not so caused by morally free agents.  Prima facie natural evils might include the suffering of a young child dying of bubonic plague, or alternatively the suffering of a fawn dying of burns caused by a forest fire that was started by lightning.  It is these prima facie natural evils that tend to cause the most difficulty for belief in the existence of omniGod.  As such, a useful approach to considering this issue is to constantly bear in mind just such cases of suffering, and try to reconcile this extreme suffering with the actions of an all-powerful and all-good god.

So, why might one hold that evil, understood in this way, doesn’t exist?  One way is to say that we are flawed, imperfect creatures, and, as such, we have a flawed, imperfect understanding of the world and of what is good and evil.  Perhaps we mistakenly think there is evil, when, in fact, there is none.

One counter-reply to this is to say that surely, at the very least, the illusion that there is evil is itself a form of evil.  It is surely a horrific state of affairs in which an all-powerful being would allow us to be so deluded that we think we are suffering so extremely when, in fact, we are not.  How could an all-powerful morally perfect being allow us to be so badly deluded?

Another atheistic counter-reply is to point out that most, if not all, theistic religions readily accept the existence of evil in the world, and further, they usually rely on this to explain the need for human salvation (if there is no evil, then there is no need for salvation).

In other words, one would be very hard pressed to maintain that evil, understood in this way, doesn’t exist.  It seems all too evident that it is real.

 

A second preliminary theistic response

One way of denying the logical incoherence of the three claims turns out to be very similar to the denial of (3) above.  It grants that evil exists and that God is all-good.  But it takes seriously the claim that God’s ways are not our ways.  More specifically, it is the thought that God’s idea of good and evil is not the same as our idea of good and evil.  God’s goodness is consistent with there being evil (as we understand it) because God’s goodness is of a different kind altogether from our goodness, and it is not just greater in degree than our goodness.

The atheistic counter-reply to this is to say that we then ought not use the word “good” to describe God, since we have no right to call different things by the same name.  The theist who gives this reply is, contrary to appearances, not an omniGod theist.  As John Stuart Mill emphatically says in Mr. Mansel on the Limits of Religious Thought:

 

I will call no being good, who is not what I mean when I apply that epithet to my fellow creatures; and if such a creature can sentence me to hell for not so calling him, to hell I will go.

 

Thus, I don’t think this line of theistic defence works, if one wishes to stick to omniGod theism.

 

A more detailed version of the Argument from Evil

The more serious theistic attempts at replying to the Argument form Evil are those that similarly deny the incoherence of holding all three claims true, but don’t do it by muddling definitions.  In this way they aim to deny the strong, logical version of the Argument from Evil.  How do they argue that holding all three claims is coherent?  To answer this, it is useful to give a more detailed formal version of the argument.  This allows us to set out more precisely the assumptions being made on both sides, and allows the theist to articulate more precisely which of the atheists’ assumptions s/he is denying and why.  One such argument, structured as a reductio ad absurdum, is given below:

 

1.                  There exists a being called God [reductio] such that:

2.                  God is omnipotent (all-powerful) [definition] and,

3.                  God is omnibenevolent (all-good) [definition]

4.                  If God is omnipotent then he can prevent all the evil he intends to prevent [from 2]

5.                  If God is omnibenevolent then he intends to prevent all the evil that he can prevent [from 3]

6.                  There exists a being, called God, who can avoid any evil he intends to avoid, and intends to avoid any evil he can avoid [from 1 – 5]

7.                  Any being who avoids any evil it intends to avoid, and intends to avoid any evil it can avoid, avoids all evil [self-evident]

8.                  There exists a being, called God, who avoids all evil [from 6, 7]

9.                  There is no evil [from 8]

10.              But, evil exists [from experience]

11.              Therefore, there does not exist a being, God, who is both omnipotent and omnibenevolent. [By reductio]

 

The most common point at which the theist questions the argument is at premise 5.  The thought is that God may have a morally sufficient reason for permitting (or even directly causing) evil, which is consistent with his omnibenevolence.  It is by giving just such a morally sufficient reason that we may consistently hold that an omnipotent and omnibenevolent god exists despite there being evil.

Is this a sufficient theistic reply?  No, because the atheist may respond that it is just not possible for God to have a morally sufficient reason for allowing evil.  Such reasons are only available to agents with limited power.

So, to reply adequately, the omniGod theist must not only say there can be such reasons, but produce possible theories about what God’s morally adequate reasons for permitting evil could be.  Such theories are called either “theodicies” or “defences”.  Some writers use these terms interchangeably, while others like to say that a theodicy is a stronger claim than a defence.  A defence is a mere possibility, whereas a theodicy is a claim to know what God actually did.  Those who talk of defences point out that we should not be claiming to know the mind of God.  Those who use the word “theodicy” say that we were never intending to, anyway.

To meet the charge of incoherence, the theist merely needs to give a logically possible morally sufficient reason for God permitting (or directly causing) evil.  However, it is at this point that another version of the Argument from Evil surfaces.  This is the probabilistic version, and it is the claim that while we might grant that there is no logical inconsistency, the sorts of claims that the theist must make in order to refute the logical version makes the theistic position extremely implausible.  To meet the further charge of implausibility, the logically possible morally sufficient reason that the theist attributes to God must also be plausibly true.

But this then brings up the possible worry that what one person may claim to be implausible, another person may claim to be plausible.  This is certainly a potential problem, and while it may be possible to further examine the positions on both sides for faults in argumentation, it may eventually end up that the theist and atheist are both engaging in the argument from different, though equally coherent and rational, worldviews.  If this eventuality is indeed that case (though there is a lot of work to do to show that it is the case), then it may get to the point where no further dialogue between theist and atheist is possible – they will have to acknowledge the other’s position as rational, given the evidence, though say they don’t themselves find it plausible (or rational from their own worldview).

But this is getting ahead of myself.  The next step is to examine the various putative theodicies that theists have put forward.  If at least one such theodicy is logically possible, then the logical version of the argument is refuted.  If one is plausible, then the probabilistic version of the argument is refuted.

 

The theistic Grand Strategy

While throughout the history of this argument there have been variously many responses, the best ones all seem to fit the form of what John Bishop calls the Grand Strategy.  The Grand Strategy is to argue:

 

1.                  That even an omnipotent being cannot do the logically impossible.

2.                  That some good things are logically unobtainable unless certain evil things exist.

3.                  That the value of these good things could be so great that it would be morally justifiable, for the sake of obtaining them, to permit, or even directly cause, the evil things that are logically necessary for their existence.

 

Claim 1 is controversial, and some people think that it puts into question God's omnipotence.  However, it seems to me that an omniGod theist who is wishing to engage in the conversation and reply rationally to the atheistic claim must accept claim 1.  Without it, no theodicy can be provided, and the Argument from Evil is successful.  In addition, many people say that logical consistency is not such a big restriction, or really a restriction at all.  After all, if something is conceptually incoherent (e.g. a square circle), it doesn’t seem fair to require any being (including an omnipotent one) to be able to create it.  The alternative to accepting claim 1 is to turn away from rationality (essentially opting out of further debate), but then one would also be turning away from a purely omniGod claim (perhaps, for example, towards a negative theology).

(Parenthetically, it seems to me that many people who call themselves theists would readily acknowledge at this point that they are not omniGod theists, either by holding a “his ways are not our ways” view, or by, when pressed, rejecting claim 1.)

There are two main ways of developing this Grand Strategy.  This is either by saying that evils are a logically necessary means of obtaining the greater goods, or alternatively that the evils are a logically necessary side effect of the greater goods.  But before examining these, I wish to briefly look at what sort of theistic reply this excludes.  I can see at least two mistaken theistic responses that pretend to be forms of the Grand Strategy.

The first is that there is a logical connection between the respective conceptual understandings of goods and evils.  There are various formulations of this. One version is to say that evil must exist because if it didn’t then we wouldn’t understand what good was like.  Another is that evil must exist because if what we call evil now didn’t exist, then there would still be evil because we would call the lesser good things evil.  With respect to the first formulation, I take it that this is simply wrong – it is possible for something not to exist and yet we still have a conception of it.  And further, even if it was correct (which it isn’t), we might still want to say that a world with no evil and our being unaware of it is a better world that one in which there is evil and we are aware of it.  With respect to the second version, this just seems to be a case of changing our use of words, and all we need to say is that would clearly be a better world than this world, so why didn’t God create it?

The second response is more sophisticated, but equally mistaken.  Its mistake is to suggest a merely a contingent (causal) connection between the goods and evils, rather than a logical connection.  I take the various ‘soul-making’ responses (e.g. John Hick), to be prime examples here.  The thought is that this world is an intermediary world or a training world that prepares us for when we go to the next world (heaven).  It is in this world that our souls or characters develop and grow in some way (presumably through suffering).  Those who are successful are allowed into heaven, because presumably without this preparation we would not be able to appreciate or understand or survive in heaven.

The atheist’s reply to the soul-making idea is to say that this training is merely a causal connection, not a logical connection, and the theist is momentarily forgetting that God is all-powerful.  If there is a world without evil (heaven) in which we can exist, then why didn’t God create us already there, with our souls already prepared?  Or why didn’t he allow us to develop these character traits in other ways that don’t require the existence of evil?  There doesn’t seem to be any logical contradiction in God using either technique, since on the theist’s soul-making story, all that has happened is that the souls grew because of experiencing some causal events through earthly life.  Surely, at worst a better world would have been that God created us with experiences of this world, but did not force us to actually experience this world.

The theist must counter-reply by explaining why, contrary to the atheist’s reply, there is a logical connection between the spiritual development through actual evils and the greater good.  But, at least so far, I have not seen this done.

Another atheistic response is to wonder what the value is of these character traits that the soul has developed, since in heaven there is no evil, so one has no use for them anyway.  So, now I turn to what I think are better theistic responses that properly do follow the Grand Strategy.

 

Evils are a Logically Necessary Means

This category of theodicies states that some evils are a logically necessary means for obtaining some kinds of goods.  For example, no one can experience the joy of overcoming adversity unless there was some adversity to overcome.  The thought is that the joy in overcoming the adversity is a greater good than the evil of the adversity, so God creates the adversity in order to allow us to experience the joy.  More generally, the claim is that the second-order goods that are logically connected to the first-order evils outweigh the first-order evils, and thus God puts up with the first-order evils.  This is called Absorption Theodicy (the idea is that the second-order goods absorb the first-order evils).  (Note: if a soul-making theodicy can be shown to have potential, it would be another type of Logically Necessary Means Theodicy.)

Atheists have put forward a number of different ideas as problems with Absorption Theodicy.  I will briefly mention three:

(1) J. L. Mackie suggests that there are second-order evils that parallel the second-order goods.  That is, when the theist postulates that first-order evils give rise to second-order goods (such as that pain gives rise to courage or joy), the atheist says that these first-order evils also give rise to second-order evils (such as that pain also gives rise to cruelty), and further, these second-order evils cancel out the second-order goods in such a way that God ought not have put up with the first-order evils.  But the theist may then reply that there are further third-order and higher-order goods that outweigh the second-order evils (such as the second-order evils giving rise to the third-order goods of hope that things will ultimately turn out for the best).  The atheist may reply that there are corresponding higher-order evils (such as second-order evils giving rise to the third-order evil of despair and nihilism). More work is required than I am giving it here to examine which side wins out, but I suspect that ultimately this will end in an impasse due to epistemic limitations – that is, we just cannot know which outweighs the other.  But if this is so, and since in this context the burden of proof is on the atheist, Mackie’s response does not suffice, so we must look to see if there are better atheistic counter-replies.

(2) As I see it, a better objection is what might be formulated as a Kantian objection.  The claim is that Kantian ethics tells us that we should not use people merely as a means to an end, and it seems that, for this theodicy to work, God would essentially be using people merely as a means to an end.  Though this is not a knockdown argument, since the theist may then claim to be a utilitarian rather than a Kantian (or maybe that God is a utilitarian but we ought not be), this may still be a problem for some theistic traditions.  Some people have thought it strange that theists generally reject utilitarianism, yet, in the case of the Argument from Evil, they will often (implicitly) accept it.

(3) But even if the various problems put forward by the atheist can be overcome and this theodicy works, it has only worked with respect to some forms of evil.  That is, it explains away absorbed evil, but presumably not all evil can be absorbed.  Thus, there are still unexplained evils (those that are unabsorbed), and the theist also needs to use other theodicies.

 

Evils are a Logically Necessary Side-Effect

This category of theodicies states that evil is a logically unavoidable side-effect of something so good that a morally perfect being would be justified in putting up with it.  The most important theodicy in this category is Free Will Theodicy.  This is the view that supreme good is only possible in a world with morally free agents, and God cannot logically stop their evil actions (moral evil) without taking away their freedom.  Thus, evil is a result of morally free beings, and God puts up with their evil deeds because their freedom is the greater good.

The theist may either claim that free will is itself the greater good, or upgrade Free Will Theodicy to the claim that something else is the greater good, and this ‘something else’ is logically connected to the free will.  One example of this ‘something else’ might be, as Bishop suggests, that mutual mature loving relationships are the greater good, and these cannot come about unless the participants have significant freedom.

Again, there are a number of responses available to the atheist:

(1) The atheist may argue that free will does not exist.  That is, s/he may say that hard determinism is the correct response to the metaphysical issue of free will.  The theist will have to defend either compatibilist freedom or libertarianism.  I will not engage in debate at this point, except to say that, as a believer in compatibilist freedom, I do not think that this atheistic reply works, but I acknowledge that this is still very much a live issue.  Similarly, the atheist may claim that free will, or the “something else” that is logically connected to it is in fact unimportant.  Again, the theist will have to deny this, and I suspect that talk of mutual mature loving relationships would suffice for the theist.

(2) Much of the more recent debate on this topic centres on the atheistic counter-reply that states that, in fact, God could have produced a world with significantly morally free agents who always freely do what is right, but this didn’t happen, so he must not exist.  There are numerous atheistic theories out about how God could have done this, for example:

(2a) The Coercive/Persuasive Strategy:  The idea of this strategy is that, while God gives moral freedom, he also makes it very clear to the free agents what the difference is between right and wrong, and may further make it clear what the consequences are for such actions.  He may provide revelations about ethical laws and heaven and hell (reward and punishment), give us a moral conscience, and/or provide us with the cognitive means for analysing right action.  One problem with this is that even if it was possible it is not clear if this would succeed, and further, maybe God did try this strategy.  That is, it is not clear that even if this happened, that people would always freely choose to do the good.  Also, one might wonder at the worth of freedom if God was too heavy handed regarding dishing out rewards and punishment.

(2b) Playpen Strategy (Steven Boer, David Lewis):  On this account, there is moral freedom, but God intervenes occasionally to block the harmful consequences of wrong actions.  That is, he prevents evil through “coincident miracles.”  Again, in reply, one might wonder at the worth of such freedom.  While freedom itself might be preserved in this strategy, this freedom is not valuable freedom, so the theist may claim.  The truly valuable type of freedom might only be valuable because it gives rise to the real greater good, as I suggested above in the upgraded version of Free Will Theodicy.  Also, the theist may give an objection similar to the Ordered World Theodicy, which I shall discuss below.

(2c) Selective Freedom Strategy (Lewis):  If we assume that Incompatibilism is true, and God has middle knowledge (Molinism – his omniscience includes knowing what all possible free agents will freely choose in all possible worlds), then he could grant freedom selectively – just in the cases where the agent freely chose the good.  He would remove the freedom in just those cases that he knew that the agent would choose the wrong action.  One reply by the theist is once again to wonder about the worth of this type of freedom.  Another is to wonder if, when God withheld freedom in one instance, this would induce the free agent to commit evil on the next occasion, so God would have to withhold freedom again, and so on, and in the end God would be withholding freedom on all occasions.

(2d) Determinism Strategy (Mackie):  The idea here is that God could have set up a deterministic world in which we have compatibilist free will (Soft Determinism).  Mackie claims that there are some deterministic possible worlds in which God gave us free will and yet we always freely choose the good (Mackie-worlds), and that God ought to have actualised these worlds.  One theistic reply is to talk of what Alvin Plantinga calls transworld depravity.  This is the idea that all individual essences will commit at least one moral evil in all possible deterministic worlds.  I side with Mackie on this, that I don’t see why transworld depravity needs to be the case.  Also, if we bring in further Christian claims, we might say that Jesus is one such individual essence who it is claimed does not suffer from transworld depravity.  However, I think a better reply by the theist is to once again refer to the upgraded version, and say that if God made a deterministic world in which we all freely chose the good, he would be contriving things to such an extent that we could not engage in a loving relationship with him.

I will not go into this any further, but needless to say, this is a hotly contested area, and there is ongoing debate between theists and atheists as to whether God could have produced beings with meaningful free will, yet they do not commit moral evils.  At this point, it sees to me that all of the atheistic replies may ultimately fail, if the upgrade can be shown to be secure – that God could not have done any of these strategies, because it would have interfered too greatly with our ability to engage in a personal relationship with him, and all our relationships would be too contrived (but then one may wonder, as Bishop does, whether there is any world that God could have created which would not have suffered from this problem, and it may follow that the upgrade idea is an atheistic Trojan horse).

(3) Another major problem is that Free Will Theodicy does not explain natural evils (evil not partly caused by the wrongdoing of morally free beings), and so, like Absorption Theodicy, it must be used in combination with other theodicies.

So, the atheist may claim that even if we accept that these two theodicies above work (and some atheists do not), there are still some unabsorbed natural evils that have not been adequately explained.  One approach by the theist to this assertion is to deny the existence of these evils, arguing that they are all really moral evils or absorbed evils.  In the case of the former, some theists have attempted to give an Extended Free Will Theodicy to suggest that, in fact, all apparent natural evils are either produced by fallen angels (e.g. Plantinga) or come about as a result of Original Sin (e.g. Eleonore Stump), and are thus really moral evils.  The atheist will most likely respond with the charge that this seems ad hoc and implausible.  In any case, this would only work for a certain type of theist who is prepared to accept a fallen angel theory or who genuinely thinks that Adam and Eve's eating the fruit caused volcanoes, floods, viruses, and the like.  While some theists are prepared to defend this line of thinking, others do not, and, to a large degree, the plausibility of various other related metaphysical issues come into play at this point.  And, as far as I am concerned they are very implausible.  However, at least some are probably not impossible, and hence the logical version of the Argument from Evil would seem to fail at this point.

Another approach to unabsorbed natural evils is to develop another “side-effect” theodicy that shows that the world must be a relatively well-ordered place.  This is called Ordered World Theodicy.  The theist’s claim is that for morally free agents to have meaningful free will they must have some idea of the sorts of options and consequences on offer, and thus the world must be a relatively well-ordered place.  If God had have set up a world without order or alternatively always intervened to stop these evils, it would have made the world too chaotic and unpredictable for agents to live with meaningful free action.  The claim is that there is a logical connection between an ordered world and the existence of certain natural evils.  Like the Free Will Theodicy, this theodicy rests on the assumption that free will (or whatever requires it) is indeed a greater good.  Also, it rests on the assumption that God could not have set up a world in which the laws resulted in less potential for suffering and that God could not have miraculously intervened any more than he does now without effecting the meaningfulness of the free agents’ freedom.  These (especially the latter) might turn out to be dubious assumptions to make, and the atheist will claim that God could have set things up differently or intervened more often, and, contrary to the theist, this would not have destroyed the meaningfulness of the agents’ freedom.

 

Conclusion

As I see it, once we get to this point we are, like in the cases of the apologetic arguments for God’s existence, heading towards a deep impasse between the theist and the atheist.  The atheist is going to continuously maintain that the theodicies don’t work and are unable to account for the amount and extent of suffering that exists in the world.  The theist is going to maintain that all evils can be explained, or are explainable, even if we don’t know precisely what God’s reasons were.  To the extent that I think that it is possible that all evils can be explained, I side with the theist.  In this sense I think that the logical version of the Argument from Evil fails.  But to the extent that I think it is plausible that all evils can be explained, I side with the atheist.  In this sense I think that the probabilistic version succeeds in creating a lot of doubt in the existence of omniGod.  That is, while I think that some (but not all) of the explanatory stories that the theist gives in reply to the Argument from Evil are coherent, I think they rest on dubious and implausible (though not necessarily impossible or inconsistent) assertions about the state of the world.

So, how does this leave me with respect to the Argument from Evil?  Well, as an argument to show that theistic belief is incoherent, I don’t think it works.  I think that the theist has been able to show that it is possible that the existence of an omnipotent omnibenevolent being is consistent with the amount of evil in the world.  And if a theist further claims, based on her/his worldview, that this story is plausible (though much elaboration is required at this point), and if this worldview is coherent and consistent with the evidence, then I think that this is sufficient for rationality.  Thus, I think that theistic belief, as a whole, is not irrational (though some forms of it are), at least by the standards of the Argument from Evil.

 

9.0 Conclusions

 

(14,746 words)

 

Bibliography:

Adams, R. M. (1972). Must God Create the Best? Philosophical Review, 81.

 

Angeles, P. A. (1997). Critiques of GOD: making the case against belief in God. Amherst, NY: Prometheus Books.

 

Anselm. Proslogion. In J. Hick (ed.) (1964), The Existence of God. Macmillan.

 

Aquinas. Summa Theologica (third article). In J. Hick. (ed.) (1964). The Existence of God. Macmillan.

 

Bishop, J. (1993). Compatibilism and the Free Will Defence.  Australasian Journal of Philosophy, 71(2).

 

Bishop, J. (1993). Evil and the Concept of God. Philosophical Papers, 22(1) 1-15.

 

Bishop, J. (19998). Can there be Alternative Concepts of God? Nous, 32(2) 174-188.

 

Camus, A. (1960). The Plague (S. Gilbert, trans.). Middlesex: Penguin. (original work published in 1947).

 

Cupitt, D. (1997). After God: The Future of Religion. London: Weidenfeld & Nicolson.

 

Davies, B. Introduction to the Philosophy of Religion.

 

Descartes, R. (1968). Discourse on Method and the Meditations. (F. E. Sutcliffe, trans.). London: Penguin.

 

Gould, S. J. (1980). The Panda’s Thumb.

 

Hajek, A. (1998). Pascal’s Wager. Retrieved 13 March 2000 from the World Wide Web: http://plato.stanford.edu/entries/pascal-wager/

 

Hick, J. Philosophy of Religion.

 

Hume, D. (1963). An Enquiry Concerning Human Understanding, and Other Essays. E. C. Mossner (ed.). New York: Washington Square Press.

 

Kant, I. (1965). Critique of Pure Reason. (N. K. Smith, trans). New York: St. Martin’s Press.

 

Kenny, A. (1992). What is Faith? Essays in the Philosophy of Religion. Oxford: Oxford University Press.

 

Le Poidevin, R. (1996). Arguing for Atheism: An introduction to the Philosophy of Religion. London: Routledge.

 

Lewis, D. (1994). Evil for Freedoms Sake? Philosophical papers, 24.

 

Lycan, W. G., & Schlessinger, G. N. (1992). You Bet Your Life: Pascal’s Wager Defended. In R. D. Gievett & B. Sweetman (ed.), Contemporary Perspectives on Religious Epistemology. Oxford: Oxford University Press.

 

Mackie, J. L. (1955). Evil and Omnipotence. Mind, 64 200-212.

 

Mackie, J. L. (1981). The Miracle of Theism: Arguments For and Against the Existence of God. Oxford: Clarendon Press.

 

McKim, R. (1984). Worlds Without Evil. Int. J Phil Rel, 15 161-170.

 

Paley, W. (1836). Natural Theology.

 

Pascal, B. (1910). Pascal’s Pensees. (W. F. Trotter, trans.).

 

Perszyk, K. J. (1999). Compatibilism and the Free Will Defence: A Reply to Bishop. Australasian Journal of Philosophy, 77(1).

 

Pike, N. (ed.) (1964). God and Evil. Englewood Cliffs, NJ: Prentice-Hall.

 

Plantinga, A. (1967). God and Other Minds: A Study of the Rational Justification of Belief in God. Ithaca, NY: Cornell University Press.

 

Plantinga, A. (1974). God, Freedom and Evil. New York: Harper and Row.

 

Putt, B. K. (1997). Indignation Towards Evil: Ricoeur and Caputo on a Theodicy of Protest. Philosophy Today.

 

Robinson, T. A. (ed.) (1996). God. Indianapolis: Hackett.

 

Schlesinger, G. N. Miracles. In P. Quinn & C. Taliafero (eds.), A Companion to Philosophy of Religion, pp. 360-366.

 

Smart, J. J. C. & Haldane, J. J. (1996). Atheism and Theism. Oxford: Blackwell.

 

Sober, E. (1995). Core Questions in Philosophy. Englewood Cliffs, NJ: Prentice Hall

 

Stump, E. (1985). The Problem of Evil. Faith and Philosophy, 2.

 

Taylor, R. (1974). Metaphysics.

 

Tierno, J. T. (1999). On the Proper Formulation of the Problem of Evil. Sophia, 38(1).

 

Tillich, P. (1951). Systematic Theology. James Nisbet & Co. Ltd.

 

Tooley, M. (1991). The Argument from Evil. Philosophical Perspectives, 5.