Saturday, December 15, 2012

The End of Christianity?

You (assuming you to be a regular reader of this blog, rather than someone who has just Googled their way here for the first time) may remember that I read Chapter 5 of the book 'The End of Christianity' back in April, and blogged about it. The book is a collection of essays by various of the usual atheist suspects who write books like this, edited by John Loftus. This is the companion book to 'The Christian Delusion' which I reviewed at the tail end of last year. I've finally got around to reading the other chapters in the book...

As with my review of the earlier book, I'll give my conclusions up front, and then go through each chapter one by one. So if all you want is the quick summary, you can ignore the details.

As a couple of the chapter authors from the previous book ended up reading my review here, including the editor himself, I'll say "hello" upfront, just in case they're back.

Anyway, in conclusion:

This is a more coherent and stronger case against Christianity than 'The Christian Delusion'. The over-reliance on probability concepts in a few of the chapters and logic in the last may put off the non-mathematically-minded reader, but the arguments presented are generally strong, clear and compelling. However, this isn't a book that is likely to be read by many Christians who are open to the idea of reconsidering their faith. It comes on too strong and too forcefully and I would expect that many Christians wouldn't make it through the first few chapters if this was the first book of this kind that they had read. This is really a book for already committed 'free thinkers' to provide them with ammunition in debates with theists. I have yet to read a book like this which is actually good for getting Christians to reconsider the foundations of their faith.

While certainly a stronger book than The Christian Delusion, this book also suffers from the feeling of being a compilation of arguments which are just lumped together, but do not make a coherent whole. Having said all that, there is plenty in here that seems new and fresh, and there is an awful lot of food for thought. If you're a theist/apologist looking to hone your skills, then these are the arguments you have to defend against. If you're a skeptic who interacts with theists, there is much here to be used and cited in debates. If, like me, you're someone in the middle ground still wrestling with faith and god concepts, there is much to wrestle with here. On the whole, I haven't yet seen theist arguments as strong as some of the anti-theist arguments presented here. It leaves much to think about, which is probably the best endorsement of such a book.

And now on with the full review:

In Chapter 1, "Christianity Evolving: On the Origin of Christian Species", Dr David Eller retreads much of the same ground he trod in the opening chapter of The Christian Delusion. As I said, I found that chapter a peculiar opening to that book, whereas this chapter works better as a book opener and presents its case in a clear and compelling way. The point is this: there is no such 'thing' as Christianity - it has developed, mutated, branched and evolved through its history in a manner similar to biological evolution. Christianity today bears as much relation to Christianity in the first century as contemporary Homo Sapiens bear to the earliest mammals who cowered in burrows while dinosaurs ruled the earth. The narrative told in this chapter is compelling, plausible and coherent. It could be that Christianity developed in the ways the chapter says. The problem I found with this chapter is the lack of evidence presented. The theory is compelling, but it is not demonstrated using much evidence and seems a bit lacking in citations. I'm convinced by the reasoning presented, but maybe because I have the will to believe it, but no entrenched Christian believer will be swayed by this as the evidence presented is simply not strong enough.

Chapter 2, "Christianity's Success was not Incredible", by Dr Richard Carrier, is a much stronger case against Christian belief. This is clearly a summary of the material presented by Carrier in his book "Not the impossible faith" (which I have yet to read), and the self-citations and frequent comments to the effect of "I have proved this point elsewhere, so don't need to do it again here" are mildly irritating. But aside from that, this chapter makes a strong case for the mundane nature of the Christian religion. Christianity had an unremarkable growth rate which is similar to other religions; other religions followed dead saviour figures; Christianity became - in essence - a domesticated variant of Judaism, so pagans could convert to the Jewish religion without having to sacrifice tender body parts or adhere to any set of dietary laws; and so on. Carrier convincingly shows how the set of early Christian beliefs made for a religion which could be popular and grow at the rate it did, with no divine involvement.

The most interesting bit of the chapter for me was where Carrier showed that the emphasis Christianity puts on faith over evidence explains a lot. Conversion to Christianity is based on feelings, not facts. He says:
"every early discussion we have from Christians regarding their methodology for testing claims either omits, rejects, or even denigrates rational, empirical methods and promotes instead faith-based methods of finding secrets hidden in scripture and relying on spiritual inspirations and revelations, and then verifying all this by whether their psychosomatic "miracles" worked and their leaders were willing to suffer for the cause. Skepticism and doubt were belittled; faith without evidence was praised and rewarded. Its no surprise such an approach would be "successful" because such an approach is purely psychological - it does not depend on any actual evidence." 
Carrier ends the chapter with two of his usual arguments. Firstly he wheels out Bayes's Theorem to demonstrate, mathematically, that Christianity is improbable. He's written a whole book on the subject now, but that's something to read another time. I know enough about Bayes (it was the basis of my PhD, after all) to see that Carrier uses the theorem appropriately and his reasoning is sound. It convinces me that Christianity is unremarkable. However, I feel that this approach might actually blind the layman with science and cause most to switch off. Secondly, he makes his usual claim that if there was a God he should appear to all people at all times, individually, and has no good reason for not doing this. This argument will never work against a believer, so I have no idea why Carrier still uses it. The only people this will convince are those who are already skeptical. So what's the point? In some ways, I feel the end of the chapter may actually undermine the strong case he makes in the earlier parts of it. But anyway, I suppose a strong case is a strong case.  

In Chapter 3, "Christianity is Wildly Improbable", John Loftus goes for the Christian jugular. He sets out to show that most of the main claims by evangelical are simply so improbable that nobody who considers them objectively should believe in them. He starts by presenting a list of ten beliefs which I will summarise here:
  1. There is an eternal, omnipotent, omniscient, creator, triune God
  2. There is a devil, demons and angels
  3. The earth is only a few thousand years old
  4. There was a literal Adam and Eve and their actions brought the consequences of sin on everyone since them
  5. God has a morally sufficient reason for permitting evil and suffering in the world
  6. Jesus was born of a virgin, could walk on water, multiply food, etc.
  7. The Old Testament prophesied Jesus crucifixion, which was an atoning sacrifice.
  8. Jesus was raised bodily and ascended bodily to heaven
  9. The collection of 66 ancient writings contained in the bible are infallible
  10. There is life after death, only Christians get to go to heaven, while everyone else goes to hell.
Loftus claims that each of these statements is improbable, and so the more of these a believer assents to, the more and more wildly improbable their faith is. The implication being that the chances of all of those being true is so vanishingly small that there really is no grounds for believing the whole package. The chapter retreads some of the Bayesian stuff just presented by Carrier, but in a much more hand waving and less mathematical way. But his case boils down to this:
"I am skeptical of the extraordinary claim that Jesus resurrected because I cannot dismiss my present experience. I must judge the past from my present. I cannot do otherwise. [...] If in our world miracles do not happen, then they did not happen in first-century Palestine, either. And that should be the end of it."
The case is quite strong, but I feel it is using the probability of maths in a misleading way. Loftus takes a belief system and breaks it into bits, declaring each bit 'improbable' and then assuming that if you recombine the bits, the improbabilities are multiplicative and so as you combine them the ensemble gets more and more improbable. The problem with this line of reasoning is that the improbability of the end result is entirely dependent on how many pieces you break the original into. Put some numbers on it. Break the belief into 5 bits, declare them all only 1% probable and recombine - the ensemble has a probability of 0.015, which is 1x10-10. But if you broke the original belief into 100 bits and declared each of them 1% probable, then recombine, then the ensemble is 1x10-200, a considerably more improbable entity. Loftus is throwing probability about arbitrarily. OK, his case might be sound, but his methodology is wrong. Just because he doesn't do the numbers doesn't mean he's using the concepts correctly. But I'm nit-picking here.

Loftus then wanders through all religious belief systems using the same (flawed) reasoning to try and show that the more complex the belief system, the more improbable it is. So Christianity is more improbable than Deism, and so on. Again, this reasoning relies on assumptions of arbitrary levels of improbability at each stage of complexity, and again the more layers you split the system into, the more improbable it becomes. Probability doesn't work like that. And, of course, improbable events actually do occur all the time. The probability that the whole chain of evolutionary events since the beginning of time would give rise to John Loftus is so vanishingly small that it might as well be zero. Therefore John Loftus does not exist. Except probability doesn't work like that.

The chapter jumps track here and looks directly at the concept of the Trinity. Yes, its improbable too. Then chapter criticises three Christian philosophers and some of the claims they've made. Its OK to criticise the methods and conclusions of others, but when this comes just after you've used sloppy reasoning yourself, the case is undermined somewhat. The chapter seems to be jumping about at random going "...and another thing..." rather than having a coherent message. The points are all valid, but a bit scattershot. The chapter concludes with 15 more scattershot nuggets, each making a valid point, but the aim seems to be simply to hit the reader with as many 'facts' as possible, rather than making a coherent case. Which makes for an unsatisfying chapter, even if it makes good points.

The next chapter (#4) "Why Biblical studies must end" by Hector Avalos appears to be a summary of Avalos's own book "The End of Biblical Studies" (which I haven't read, but I have heard him interviewed on the subject before). His point is that the Bible is just an ancient document, no more, no less and "we should now treat the Bible as the alien document it is, with no more importance than the other works of literature we ignore every day." His point is that the Bible is not special enough to merit the special and privileged study it has received to date. Indeed he asserts that:
"the main subdisciplines of Biblical studies have succeeded in demonstrating that the Bible is the product of cultures whose values and beliefs about the origin, nature, and purpose of our world are no longer held to be relevant, even by most Christians and Jews."
He then goes on to look at each subdiscipline in turn:
  1. Translations: Here it is claimed that translators of the bible 'lie' by deliberately mistranslating the bible to make it understandable and acceptable to modern Christians. For example, the polytheistic nature of the OT is masked by translating the names of discrete gods (e.g. Elyon and Yahweh) in such a way as to suggest that these are simply different ways of referring to the same God. Other examples are given. (Although my personal favourite is not mentioned - Psalm 139v9: "If I rise on the wings of the dawn..." should really be rendered "If I fly with the wings of the dawn Goddess...", but anyway...).
  2. Textual Criticism: Here it is claimed that study of the text of the Bible has, essentially, destroyed any credible claim that the bible is inerrant or unchanging, as it is clear that the text has been changed and modified in transmission.
  3. Biblical Archaeology: This section may be summed up by the following quote from Ronald Hendel: "Archaeological research has - against the intentions of most of its practitioners - secured the nonhistory of much of the Bible before the era of kings." Avalos takes it further and shows that even much of the era of the kings is in doubt.
  4. The Unhistorical Jesus: The longest section in this chapter concerns the 'quest' for the 'historical Jesus'. Avalos declares the quest "is an abject failure. Further progress is futile because we simply don't have any preserved accounts of Jesus from his time or from any proven eyewitnesses." Thus the claim is that the (ongoing) quest for the historical Jesus is pointless and will never come to a conclusion.
  5. Literary Criticism: Here the point seems to be why set the bible apart from other ancient texts? And if there is no compelling reason to treat it differently, then we should stop now and take a good long look at other texts, because until now we have given special attention to the bible. 
  6. Biblical theology: The point here is that people and faith groups of all different varieties interpret the text to be whatever they want it to be. And this appears to be acceptable to theologians, on the whole. Thus, the bible is made to say things that the authors never intended, and thus actually studying the bible in this way is pointless.
The chapter concludes with a number of statements including these:
"Biblical scholars all agree the Bible is a product of another age and culture, whose norms, practices, and conception of the world were very different from ours. Yet these same scholars paradoxically keep the general public under the illusion that the Bible does matter or should matter."
"Why do we need an ancient book that endorses everything from genocide to slavery to be a prime authority on our public or private morality? 'The Bible' is mostly a construct of the last two thousand years of human history."
So the point is made, the actual Bible is irrelevant in the modern age, and the public conception of the Bible (i.e. the message that is perceived by many to be not only relevant, but essential) is a largely inaccurate interpretation only loosely based on the real thing.

I considered Chapter 5 "Can God exist if Yahweh doesn't?" back in April, so you can read my thoughts on that chapter in the older post.

Chapter 6, "God's Emotions: Why the biblical God is hopelessly human" by Valerie Tarico is fascinating. It looks at the character of God as presented in the bible, in particular his emotions, and examines the implications of the biblical claims. In the bible, God clearly has human emotions. He gets angry, he is jealous, he is occasionally surprised, he is defined by his love, and so on. The chapter looks at what emotions actually are and why they are integral to being human, and why they are, or should be, not at all integral to being divine. The core message of the chapter is summed up in this quote:
"If I asked you whether God has a nose or a penis, what would you say? Most Christians would say probably not. A nose is for breathing and smelling. A penis is for sex and peeing. God has no need of either. In the same way, I would argue that God has no need for emotions - intricate chemical reactions designed to activate and direct bodily responses to the external environment. As wonderful as emotions are, they are made of and for the fabric of this natural world."
There's a lot more in this chapter, particularly with regard to anger (God can either be angry or omnipotent, there is no point in an omnipotent being getting angry - what can he possibly get angry about?) and some more positive emotions, but if you want to know about all that, you should simply read the chapter for yourself. Its fascinating.

The third section of the book begins with Chapter 7 "The Absurdity of the Atonement" by Ken Pulliam. This asks the question 'Just what did Jesus achieve by dying on the cross?' and unpacks most of the most common Christian answers to demonstrate that they are all, at some level, completely absurd. I have to say that this chapter is aimed squarely at the beliefs of the 'Reformed' branch of the church, so other Christians might actually agree with some of the points made, but it is a strong attack on that, specific, theology. The chapter states that:
"The purpose of this chapter is to show that the dominant view of the atonement in Evangelicalism, the view that many claim is the very heart of the Christian Gospel, is illogical, immoral, incoherent, and therefore, absurd."
It is illogical in that it makes no sense for an innocent party to take the punishment of the guilty. It is immoral because it inflicts pain and suffering on an innocent party. It is incoherent in that the way it is expressed is that the actions of the son are able to propitiate the wrath of God the Father, but if the doctrine of the Trinity is assumed, then it must be that the Son and the Spirit also need propitiated, and how can Jesus' self-sacrifice achieve that? (There's a lot more to it than this, I'm summarising.)

There is an awful lot for the believer to wrestle with in this chapter, but the bit that I found the most interesting what the section concerning human sacrifice (within the 'immoral' section). Jesus' death on the cross is, essentially, a human sacrifice, is it not? The chapter quotes an evangelical, David Dilling, on the subject of human sacrifice in pagan religions, as condemned in the OT:
"The greater offense is not the sacrifice, but the idolatry involved in offering such a sacrifice to a god other than Yahweh. The first commandment is not 'Thou shalt not offer human sacrifices,' but 'Thou shalt have no other gods before me.' The Bible contains no prohibitions of human sacrifice to Yahweh. The only possible exception to this principle is the legislation regarding the redemption of the first-born sons in [Exodus] 13:1-16. This passage, however, does not condemn human sacrifice. On the contrary, it proves that Yahweh had a very definite claim on all the first born of Israel, whether man or beast."
Yikes! Dilling is also quoted as saying:
"The most frequent objection raised against the Biblical presentation of Yahweh and his relationship to sacrifice is that sacrifice, whether of human beings or of beasts, is an element of primitive religion, and that Yahweh really deserves not sacrifice at all but obedience. This view, carried to its logical conclusion, would eliminate the necessity of the sacrificial death of Christ. This in turn eliminates the atonement and thereby abnegates the whole Christian Gospel."
Bigger yikes! Of course, this is only one view within Evangelicalism, but I think it is a biblically consistent one, so should cause problems for the believer who actually thinks through the issues. As Pulliam says:
"Once again the Evangelical has a tremendous problem. His doctrine of the atonement is based on the religious rite of human sacrifice, which is recognized universally today as an immoral practice."
Chapter 8, "The Salem Witch Trials and the Evidence for the Resurrection" by Matt McCormick has a very straightforward case to make. It is basically this (my summary): The best evidence we have for the resurrection is that a bunch of mostly uneducated people two thousand years ago believed that it had happened. However, if you look at the Salem Witch Trials, we have immensely more evidence that far more people of better educational statuses firmly believed (and testified in court) that various accused people in Salem were witches. Today, despite all the evidence, we do not believe that there were any genuine (i.e. able to effect magic and spells) witches in Salem, so why, on the basis of considerably less evidence, do some people believe that a magical resurrection took place in Jesus' case? The case is a strong one and is presented very reasonably.

The only niggle I have here is that there is a certain degree of comparing apples with oranges. Nobody claims the Salem Witch Trials have any life-changing effects today, while millions will claim that their present-day experience of Jesus validates their beliefs about the past. The two cases are only equivalent to an impartial historian, not a believer.
In Chapter 9 "Explaining the resurrection without recourse to miracle", Robert M. Price advocates a case which he, himself does not actually believe in. That is OK, he states this up front and explains that he is granting the apologist more 'facts' about the resurrection than he believes to be valid, for the sake of argument. The point being that, even granting half of the ground of the debate to the believer, they are still not justified in their conclusion that a miraculous resurrection happened.

Price examines the 'Swoon theory' (that Jesus didn't die, but fainted on the cross and recovered in the cool of the tomb), the wrong tomb theory (Mary and the others didn't know which tomb Jesus was in and went to the wrong one, which was empty) and the mistaken identity theory (the post-crucifixion sightings all involved people other than Jesus, and the disciples, after the meetings all concluded "that must have been Jesus, even though we didn't recognise him..."; see Emmaus road story for example) and shows that there is good evidence for each of these in the text of the new testament. Furthermore, he shows how he believes each of these are more likely than that a miracle actually occurred.

He then treads some of the same ground as Carrier by showing (again) that the growth of Christianity was mundane, not miraculous. Having had this already in this book, this felt a bit unnecessary and it added little to the argument of the earlier part of the chapter.

Next we come to Chapter 10, "Hell: Christianity's most damnable doctrine" by Keith Parsons. This chapter contains very little that is new and, indeed, treads much the same ground as Rob Bell's controversial book 'Love Wins' from last year - and that was written by a Christian minister. The main points of the chapter are basically that the very concept of hell is immoral and incoherent and yet that the doctrine of hell probably explains why there are so many Christians - generations of people were scared into following the teachings of the church, because they were scared of the prospect of hell otherwise. Of course, hell is not a uniquely Christian belief, but this fact is not considered here. The chapter presents some Christian beliefs about hell (generally quoting long dead preachers and church leaders), then dissects some apologetics arguments on the subject, including those of CS Lewis. The point is well made, but as I said it is nothing new and has been made many times before. But I liked this quote from Eddie Tabash:
"...the very doctrine of hell is so horrific that it probably deters many from believing, and so condemns them to hell..." Indeed.
Chapter 11, by David Eller, starts the fourth section of the book on science vs. Christianity by asking "Is religion compatible with science?" You can probably guess the answer. The chapter first breaks down what we mean by 'religion' and what we mean by 'science'. I like the Carl Sagan quote cited here:
"Science requires the most vigorous and uncompromising skepticism because the vast majority of ideas are simply wrong, and the only way to winnow the wheat from the chaff is by critical experiment and analysis."
Strong stuff. The chapter goes on to show the occasions where Christianity has given in to science, such as the earth going around the sun, evolution (OK, that debate is still ongoing for some), and so on, and it is clear that there are not equivalent instances the other way around; science advances and Christianity modifies its beliefs. Eller sums this up by saying "beliefs are changeable, but facts are not", a statement that may not be entirely accurate as 'facts' are generally interpretations of data, and as new data emerge, occasionally 'facts' change. But the point remains, generally science changes Christianity, not the other way around. Christianity is fighting a losing battle.

Eller presents a couple of quotes that more or less prove his case, first this from Tertullian:
"After Jesus Christ we have no need for speculation, after the gospel no need of research. When we come to believe, we have no desire to believe anything else; for we begin by believing that there is nothing else which we have to believe."
This mindset has been part of the Church pretty much since the very start and still endures in parts of Christianity today. If we believe this, then science is not merely incompatible with Christianity, but irrelevant. The second quote comes from Luther:
"reason must be deluded, blinded and destroyed" and "faith must trample under foot all reason, sense and understanding"
That certainly is an anti-science mindset.

But it goes both ways. Science cannot accept the fundamental assumptions of supernaturalism. Science, essentially makes claims like Force = Mass x Acceleration. Religion must modify even the most basic scientific claims to allow for supernatural agency. So Force = (Mass x Acceleration) + (Supernatural Force). If this were the case then all of science would fall apart. But it doesn't. Science works, therefore Supernatural Force must = zero. Therefore there is no supernatural.

The next target, in Chapter 12, is 'Intelligent Design'. Richard Carrier again uses Bayes theorem / probability to show that the evidence of the world around us increasingly supports the hypothesis that the universe is the product of random processes and increasingly undermines and refutes the hypothesis that the universe was designed and created by an intelligent being. The case is strong and well presented, but again, the non-mathematically minded reader will switch off early into the discussion. Carrier's focus here is not merely 'evolution', although of course that is covered, but apparent design on all scales from the cosmological down to the inner workings of the brain.

One observation that is made, which I am sure is not original here, but I had never thought about before, concerns the 'bacterial flagellum' that Michael Behe and other ID proponents present as being 'irreducibly complex'. I hadn't really thought about the fact that this 'biological motor' is actually part of the E.Coli bacterium. As Carrier says:
"Since the flagellum Behe says must have been intelligently designed is what gives the bacteria the ability to move around, it actually greatly magnifies its lethality to humans. In fact, that's pretty much all it does - which means that's what its for. In other words, Behe is essentially saying that someone genetically engineered bacteria specifically to kill us."
Hmmm. The odd thing I thought about the way Carrier handled the probabilities in this chapter was to fix the Bayesian prior probability for there being an intelligent designer as being low at the outset. Not very low, but somewhere like 25%. Why does he do this? If a (not very mathematically literate) Christian were reading this, it would look like the argument works only if you assume a low prior probability. I think the case would be stronger if you started from a high probability prior - something like a belief that is 95% or even 99% sure that there is an intelligent designer. And then a cumulative case could be presented that would show that even from a starting point of almost certainty in God, the logical conclusion would have to be that the intelligent designer is extremely unlikely. By starting with a low prior and ending up with a lower, but not that much lower posterior, the case doesn't actually look as strong as it is.

Chapter 13 is "Life after death: Examining the evidence" by Victor Stenger. This chapter is basically a response to a paper written by Dinesh D'Souza called "Life after death: the evidence?" which I haven't read. While there is a lot of good reasoning in there, this is clearly one side of a debate and is lacking as a consequence. Stenger considers the evidence carefully and makes a good many valid points. My favourite of these is:
"If we have disembodied souls that, as most religions teach, are responsible for our thoughts, dreams personalities,  and emotions, then these should not be affected by drugs. But they are. They should not be affected by disease. But they are. They should not be affected by brain injuries. But they are."
Also the line "the plural of anecdote is not data" made me chuckle. Suffice it to say that this is a fairly robust attack on the flimsy evidence that there is for life after death or the existence of a soul which could exist outside of the body.

Richard Carrier is back again in Chapter 14, "Moral facts naturally exist (and science could find them)". I have to admit that this chapter bored me and now, some weeks after actually reading the chapter, I can't really remember much about it. Logic is a very precise way of tacking a subject and proving something or other about it, but it rarely makes for an entertaining read. I have no idea whether there are objective moral facts or not, this chapter doesn't really answer that question either, but merely makes the claim (in a thorough and logical manner) that if there are logical facts then science is a better way of getting to them than religion. Along the way I highlighted a few lines of text including this, which is a point well made:
"We must conform our beliefs to what we discover, not reject all discoveries that fail to conform to our beliefs"
This chapter comes with an appendix which bored me with even more formal logic than the main body of the chapter.

And that is that. 14 chapters of ammunition to be used in the theist vs. atheist debate. All that remains is a brief afterword:

In "Changing Morals and the fate of Evangelicalism" Robert M. Price makes a simple but very valid point. The point is this - Evangelical Christianity is dying - within a generation or two it will be gone. his main evidence for this is the observation that nothing now distinguishes Evangelical Christian youth from their non-Christian counterparts. In generations gone by Christian sub-culture banned certain things, like dancing, going to the movies, drinking alcohol, wearing certain types of clothes, having pre-marital sex, etc. For the most part, these distinctives have all gone. The last remaining one of sexual behaviour is slowly crumbling in our own day, and recent studies show that older high school and college age Christians (in the USA, I guess) are having about as much 'recreational' sex as their peers. Once Christianity ceases to be different, once it loses its 'moral high ground' it will be gone.


Mike Blyth said...

Thanks for your detailed review. I happen to be reading the same book at the moment. One question I'd love you to answer, since you have a background in probability in statistics, is the validity of this whole idea of using Bayesian probability in apologetics. Both sides do so.

As far as I understand, formal probability inevitably deals with a universe of possibilities with a known distribution. How then, is it in any way meaningful to assign an a priori probability to an event such as "God exists" or "Jesus rose from the dead?" The sample size is equal to the population size of one. It's not like we can answer the question, "Given a large number of universes, how many of them were created by a god."

As far as your comments on the validity of slicing events into thin pieces and then multiplying their probabilities, it seems to me that the main problem is that the author ignores the main requirement for that procedure, which is that the events be independent. The more one slices a natural event, the more dependencies are going to be present.

Ricky Carvel said...

Hi. You're right about the independent events thing...

But anyway, Bayes' theorem can be used to, essentially, choose between a number of rival hypotheses on a given topic, provided a prior probability can be assigned. So, in principle, it is able to take the two hypotheses:
1. God exists, and
2. God does not exist
and use evidence to 'refine' or 'update' the prior estimate of probability, to give the posterior estimate.

Thomas Bayes himself was unsure if a 'diffuse' prior could ever be assumed - that is, whether all hypotheses could be allocated equal prior probabilities (in this case, 50:50), but subsequent Bayesians (myself included) have occasionally justified the use of the diffuse prior, when no other prior is available.

My own work has always used Bayes in a cumulative manner, considering several independent pieces of evidence in sequence and 'updating' the probability estimate each time. That is, "today's posterior is tomorrow's prior" as Bayesians like to say. What I found was that after 5 or 10 pieces of evidence were considered the original prior became largely irrelevant. That is, the end result was broadly the same irrespective of whether the initial prior was 95:5, 50:50 or 5:95. Hence my conclusion that using a diffuse prior is acceptable.

Bayes theory is the ideal way to approach a probabilistic problem where the distribution is unknown. Indeed, that is its primary purpose, to attempt to model the unknown probability distribution. Like you I have some doubts about using it in a scenario where the actual probability must be 1:0 or 0:1 (i.e. either there is or is not a God, there can be no middle ground), but conceptually it is easier to think in degrees of belief:
1. I should believe in a God, or
2. I should not believe in a God
These can have assigned probabilities between 1 and 0, but the end result is basically the same.

I haven't read it yet, but if you want to see whether or not Bayes can and should be used in apologetics, you should try Colin Howson's recent book 'Objecting to God'. I don't know to what extent he uses Bayes in this book, but he has previously written an excellent book on Bayes called 'Scientific Reasoning: The Bayesian Approach'. I'll bet (high prior probability!) that he uses probability theory in an exemplary manner in this book.

Anonymous said...

Too long to have any detailed comments but thanks for the great post Ricky.

Anonymous said...

Christianity will never die out. Two forces keep it alive, two very opposite and conflicting forces, and then by a third factor.

1.) Morality. Many people are attracted to Christianity by the moral teachings of Jesus, i.e. the Sermon on the Mount. Many of these same people, of course, ultimately leave Christianity in disgust because of Paul, so they are only like students who go to church like a school for a certain amount of time and then graduate to become Deists/Theists. But their place is always taken by more people of the same category! And then there is the next category:

2.) Immorality/Libertinism. Many people are attracting to Christianity because of the immoral and libertine teaching of justification by faith alone, especially as its taught by easy-beleivists and "once saved always saved" type denominations that say you can sin sin sin sin sin and go to heaven so long as you believe in Jesus. Of course, many of these people ultimately come to realize that using Jesus' death as an excuse for immorality is no more powerful than just saying "God doesn't exist" and using that as an excuse for immorality! So, they ultimately treat the church as a school also, and graduate to become atheists.

3.) Born in the church, will die in the church. These are the people there for the long haul, because they were born into it, and they just, ya know, know that it must be true because their parents believed it. This is the group that gets offended when your denomination's doctrine questions their grandmama's salvation. "I know baptism can't be essential to salvation, because my grandma wasn't baptized and I know she's in heaven!!!!" Well, that proves it then!

Anonymous said...

"I haven't read it yet, but if you want to see whether or not Bayes can and should be used in apologetics, you should try Colin Howson's recent book 'Objecting to God'."

Bayes has no appeal to me. Once you mention maths, my brain turns off. Besides, a reasonable person doesn't really need a convoluted mathematical equation to measure probability. Nor does it make any sense to measure miracles by probability. What's the probability of someone being raised from the dead naturalistically? 0%. That's what makes it a miracle. Duh.