A Measure of Faith - Probability in Religious Thought · bblais on the web

https://bblais.github.io/posts/2019/Jul/15/a-measure-of-faith-probability-in-religious-thought/

Thanks Brian

You think that the God of the universe, if he/it exists, must be a ‘thing’ (your word) in the universe, like a particle, that can be studied by scientific methods. Such a poor view of God!<< yes, God is a "thing" at a minimum, unless you're saying God is "no-thing". That's the bare minimum distinction -- exists or not. I am not going to have a better view of God until such a hurdle is accomplished -- and why would I? If I said you had such a poor view of unicorns, when I hadn't established if they even exist and you have good reason to believe they were (at least in part) made up, would that even be a reasonable response? If I showed unicorns actually existed, with good evidence, and you persisted in your poor view then I might have a point. Until then, this seems to be a non-sequitur< You also say Either God interacts with the world, in which case the interaction can be studied, or he doesn't. If the interaction is so nebulous to be indistinguishable from nothing, then Occam kicks in and we don't have good reason to believe God exists. Even in the story, God says that he acts in the world, and wants a personal relationship. As I said, I treat this Christian response as an admission of defeat. One can see this easily when Christians very quickly jump on the bandwagon when an experiment seems to show God exists -- science justifies God! Once the experiment is shown to be incorrect, suddenly God can't be studied by scientific methods. It's a sour grapes response, given exactly the same way by other discredited claims. The parallels are striking! It has nothing to do with naturalistic bias -- I'd love there to be a supernatural realm, totally open to the evidence for it. What's the best case? I've never heard anything even remotely convincing.< No it cannot necessarily be studied. If he acts supernaturally, as would be expected, then it cannot be studied by naturalistic means, with experiments etc. There would be no regularity or predictability to study.

See below also on this confusion between natural and supernatural, where we talk about what it would take for you to accept a miracle and you say it has to be understood in naturalistic terms. Well then, it would not be a miracle, would it?

And suppose he does not interact at all, being either the deist god or the source of all being? He exists, but likewise cannot be studied! In that case you say we don’t have good reason that God exists. This is your naturalistic mindset. Consider the case of the deistic god, who designed the universe and then let it run without intervening. Suppose he exists, for the sake of argument. He does not interact, and yet the evidence is all around us in the form of the design of the universe. Are you really saying that he left no good reason to believe he exists?

Be that as it may, the main point is that the creator of all, if he exists, cannot be one of his own creations, cannot be ‘a thing’ in the universe. The sustainer of all cannot also be one of the things sustained. Here are two reviews of a book written 4 years ago by Rupert Shortt, to give an idea of what I am saying. https://www.theguardian.com/books/2016/mar/24/god-no-thing-rupert-shortt-review-response-new-atheism https://www.telegraph.co.uk/opinion/2016/03/26/god-is-no-thing-but-he-is-in-charge-of-things/ Here also is the Unbelievable podcast discussion on the book. https://www.premierchristianradio.com/Shows/Saturday/Unbelievable/Episodes/Unbelievable-Have-atheists-got-the-wrong-God-Rupert-Shortt-vs-Jeremy-Rodell if you are interested to listen

God does not make it easy for people to know and worship him. Isaiah says ‘Verily thou are a God that hidest thyself, O God of Israel, the Saviour’ (Is 45:15) And what right has the creature to cross-examine and study the creator?<< Oh, a book which is trying to support a possible fantasy has mechanisms to suppress doubt? How surprising! Again, this is what the psychics and fakes all say, and again it is a dodge brought out when evidence isn't forthcoming.< Depends what you mean by evidence, doesn’t it? God is not a thing that you can put under the microscope. Do you really imagine that he is such, that you could test his works and existence like that of a (eg) a fruit fly? But Psalm 19 says that the heavens declare the glory of God. Have you ever wondered why there is anything at all, where all this existence comes from?

I meant that there is not enough info for us to be able to resolve our difference. Suppose all we have is Papias’ statement, and I choose to accept it as true and as applying to our gospel of Mark and you choose not to accept it, then I say there is not enough information for us to agree.<< I see. This happens in cases where the posterior is too sensitive to the prior. This occurs if either the prior is particularly sharp (which in these cases represents pre-existing bias toward a conclusion) or where the data is simply insufficient. The only solution is to get better evidence -- either to support the claim itself, or to justify a sharp prior. If there isn't any, then strong belief is not warranted. Convergence only happens with sufficient data. Lack of convergence need not be a failure of applying probability properly, but can simply be due to lack of data< Yes exactly. And this is true about most statements about the distant past, where by ‘distant’ I mean that the only sources are written, and there are no more available. No eyewitnesses to questions, no more written sources, no DNA evidence, no videos … So differences in probabilities about things like Mark, Jesus’ resurrection cannot be reconciled by probability theory. We are stuck with the evidence we have, and we don’t have a way of reconciling. For instance Jesus’ alleged resurrection. The atheist alternative hypothesis of grave robbing plus women and Peter experiencing hallucination plus disciples wanting to believe it is just about plausible I think. But there is no way of adjudicating between it and the resurrection hypothesis. Even if we agreed on the Bayes’ factors, we would be stuck on the difference between our priors. I being a theist would have a high prior of resurrection, say 0.6. You being an atheist would have say 0.00001! No way our probabilities would converge.

Let’s be specific. Suppose for a random miracle, not necessarily Christian, my P(M) is 0.3. Yours is, say, 0.00001. How are they going to converge?<< there are a lot of miracles, so enough verification of miracles in general (and Christian miracles specifically) could -- and should -- allow for the convergence of all but the most dogmatically biased priors. Of course, we just don't have that verification for any of the miracles.< Yes but you would not concede that any miracle has been ‘verified’, would you? According to your previous statements, for verification you want to identify the mechanism and have proof of the existence of the agent. This, for a naturalist, is impossible (as I have already pointed out). So, short of a change in priors (eg me becoming an atheist or you becoming a believer) there would be no convergence.

We just cannot conclude with certainty just on the basis of written evidence that event X happened 2000 years ago. Something last week, maybe, but not 2000 years ago.<< Even last week, we can't achieve for certain (i.e. P=1). We can be very confident in a claim from last week, but even there (depending on the claim) we still might not be able to be confident due to lack of data< I don’t think confidence is the same as P=1, is it? I would say that P(Trump supporters invaded the Capitol building on 6 Jan 2021) is not quite = 1, and you agree. But we can still claim certainty can’t we?

Rules of reasoning

consistency with standard Boolean logic
Internal consistency: If a conclusion can be reasoned out in more than one way then every possible way must yield the same result.
take into account all of the information provided that is relevant to the question.
Equivalent states of knowledge must be represented by equivalent plausibility assignments.

a) Why is it a bad thing if somebody does not follow these rules?<< Really? Try to imagine cases where it is a good thing for someone to break one of the rules above -- violating deductive logic, being inconsistent, ignoring some information, etc... Perhaps these can't be proven to be universally good, but I am fine with taking them axiomatically.< Yes absolutely we cannot prove them to be desiderata for all reasoning. My point is that you have nothing to say to someone who does not use them, beyond “You are being irrational”. He might say to you, for instance “So what if I am being irrational according to your definition? I believe that Adam and Eve were the first humans, and I choose to ignore evolutionary theory”. You have nothing to say to him, do you? NB I am not commending the believer in Adam and Eve, but just pointing out the obvious, that you cannot enforce your standard of reasoning onto others (unless you are a worldwide dictator with power to read peoples’ thoughts!)

b) Seem undetermined still. How does the method work to get our probabilities of Mark being Peter’s gospel or Jesus resurrecting to converge?<< See above about data and convergence. Also, in most of these cases, specific numerical values may not be possible -- and I am skeptical of all of the cases where they are used, from Swinburne to Carrier (although Carrier at least uses inequalities, which is a bit better).< But if we see above, we see that we seem to agree that convergence is impossible if there are not enough data. And this is true about Mark’s gospel and the resurrection. There is not enough information. Our probabilities will not converge, not even if you and I were in a room in a library with all the evidence at hand, and told to stay there until we reached agreement (unless coercion was used eg told we will not be given anything to eat until we agree!)

People may ‘take into account all the info’ in different ways!<< Actually, this is probably the result of different background info that isn't stated. I have, for example, read more info in computational neuroscience and you have read more in some other fields, which we unconsciously sometimes bring to bear. That's why it is so important to hash these out -- why do you think Papias is reliable, at least in the Peter informed Mark department, for example? It is by hashing out the specifics that we might actually see all of the info, some of which we were unaware of, and thus adjust our beliefs. Sometimes it just doesn't work -- it's an imperfect system, but something to strive towards< Even if we both have all the info in front of us, we could treat it differently, put different weights on different parts, according to our own background knowledge and preferences. It looks like you are (now) agreeing with me. There is not an automatic process of convergence.

I was talking about the distinction between a law of nature, which is something about the present and the future, and a historical statement, which is only about the past – that something happened. (By convention history is only about human society, not heavenly bodies.)<< Except for astrology, which believes that heavenly bodies are tied to human societies. Also, in studies of the origin of our solar system, this is only about the past -- historical or scientific? It's so easy to come up with cases that violate this simple distinction that I just don't see the value in thinking this way< I don’t see your point, and I don’t think you get mine! A ‘study of the origin of our solar system’ is not a law of nature, surely? It is not actually any statement at all!

8.

P(ace drawn at random from a pack of cards) = 1/13 is a mathematical statement<< how do you define "at random"?< One way is to say that all outcomes, in this case, all 52 cards, are equally likely, ie that their probabilities are equal<< how would you know this is a true statement? you can define it, as a mathematical entity, but once you say "drawn" or are talking about a real pack of cards, not just an abstraction, the definition falls apart< You can define it all mathematically, using functions and sets. No we are not talking about a real pack cards, it is all abstraction. And probabilities in this simple sense are all unconditional. Conditional probability and Bayes’ Theorem came later.

Why is it only science that connect maths and logic to the real world? How about simple numeracy, like a farmer being able to count his flocks or weight his crops? Accounting? Banking and finance? These all make use of maths and connect to the real world, don’t they?<< Counting is the most basic form of quantification, and in the general sense, it's a very basic science (e.g. the farmer is doing this and applying implicit conservation rules to make sure his flock is constant, etc...)< You are defining science as anything that applies mathematics to the world. But then by definition you are right. However you have moved away from science as ordinarily understood. I thought by science you meant modern empirical science, as practised in the West since c1500. But in your new definition, the hunter-gatherers were using science when they counted how many nuts they have! Next time you redefine a word, warn me first.

10.

Why do theists constantly make this error, inserting words that I didn't say. I never said, or even implied, "confirm beyond all possibility of doubt" or "disproved absolutely" -- this is a sneaky strawman. All I want is good evidence. Get me a posterior for God above 75% or so and I'll believe, get it above 95% or so and I'll believe strongly. I do this for all claims. Just waiting for good evidence<

Moreover, you have not answered my question what, in your view, a proof of the existence of the Christian God would look like, in outline.<< I did say, with at least my outline of what would at least move the needle on miracle claims. If there was something that was repeatable, or truly vexing showing design, that would at least point in the direction of further study. A biologist once quipped about what it would take to disprove evolution, and they said "rabbit bones in the pre-cambrian". One can imagine many things like that, which would at least be an interesting puzzle and point to possible agency. If DNA showed evidence of special creation, if geology confirmed the flood, if miracle claims occurred in a way that were truly hard to ignore (e.g. amputees) then we'd at least have something to look at -- may be aliens, magic is real, or a God but at least something. Speculation past that point is fruitless, because the follow-up studies would be needed, but none of these things are occurring. Nothing even close to these sorts of things are occurring. What is occurring is theists making claims that these things are real (e.g. flood, special creation, miracle cures) for hundreds of years only to have them one-by-one shot down when we figure out that they are not correct. We did this by assuming naturalism, until otherwise shown, and we found only natural causes. This isn't like the metal detector analogy -- only finding metal. Methodological naturalism awaits someone to demonstrate supernatural agency, but until then rides on the centuries of success ignoring it. "I had no need of the God hypothesis" -- Laplace.<

Or are you consciously demanding something that you know is impossible to provide?<< Did you mean unconsciously? Neither, I think. I am looking for any evidence even suggestive of a God, to follow and be confirmed. Instead, I get stories that are so easily explained with other simpler things<

Again you ask for something repeatable, showing you do not yet understand the difference between natural and supernatural (see above). But the rest of your answer shows we are making progress. Can we be even more specific? What would move YOUR (not anyone else’s) needle for belief in (the Christian) God above 50%? Please try, it would help me a lot in my understanding of how atheists and naturalists think. Suppose eg a) The stars rearranged to spell BRIAN BLAIS BELIEVE IN ME YAHWEH one night and then returned to their original position the following. Or b) You personally witnessed a healing miracle, like the one in Times Square where a man’s legs reattach to the body after someone prays in the name of Jesus. Or … Please tell me.

11.

I don’t see the difference, in principle, between Caesar dying in old age and Caesar flying. Both have non-zero probability.<< We don't believe either, because we have good historical evidence (from many sources) that Caesar died around a particular time short of his old age. The priors for his existing as a person, of people in his position having assassination attempts that sometimes succeed do not undermine the conclusion. Even the non-negligible prior of him living to old age, which many in his status did, is easily overcome with the documentary evidence. The prior for him flying is so low, there is no historical documentary evidence that could rescue it. It's the same reason we don't bat an eye when someone has two-pair in poker two hands in a row, but are skeptical when they produce two royal flushes to hands in a row< I think we agree on this. There is no difference in principle, only difference in the values of probabilities.

I have also asked you to consider the difference between an alleged event 2000 years ago and one last week. What is the logical difference? There is a difference in evidence, because for the Times Square example there is a) your own memory if you were there b) eyewitness evidence c) video recordings.<< Not sure what you mean by "logical" difference. Sure there is a difference in quality of the evidence, the potential for corruption of the data, and in the possibility of follow-up measurements.< Exactly, no real difference in principle, just difference in quality and amount of data.

12..

How is ‘needing to establish that the agent exists’ different from ‘proving the agent exists’?<< I never said "proof of the action" or "proof of the existence of the agent" Have you forgotten what you said?<< Perhaps I have -- it's been a long conversation, and I could miss something. I did a text-search for the word "proof" in our conversation, and found only 1 case where I used the word -- and it was in the strict Boolean sense of a mathematical proof (the only proper use of the term) and not in the context of a "proof of the existence" of God. I found many uses of the term by you. Please show me the specific quote which I missed. If you can't, I'll expect an admission that you misquoted me.< I only use the word "prove" in the logical sense, as in derived from an axiomatic system. From a set of axioms we can prove (deductively) conclusions. Everything in science in inductive -- there is no "proof". I find it a particularly useful distinction because theists love to use the word "prove" to then accuse the atheist of being unreasonable, wanting evidence to remove all doubt.< OK, but I think you should have been clearer, when you give words meanings different from the usual. Most people would think ‘prove’ means much the same as ‘establish’, don’t they? But in any case, my point remains. You demand that an alleged supernatural event be demonstrated as a natural event before you accept it, a contradiction in terms. I don’t think you yet understand this contradiction in your thinking.

13..

A belief in Jesus’ resurrection, or Caesar flying, is not in itself a violation of probability theory<< Except, where it lies in a context of a scientific world, then it is inconsistent reasoning and is a violation. I was responding to modern people believing these things. Saying that a child believing in Santa Claus because that's all they've been told is not a violation of probability theory is hardly a strong point. We should know better now. Do you disagree?< No I don’t, taking your point.

14..

I thought you were saying that the NT does not contain eyewitness testimony. As for the gospels not being in first person, what about ‘This is the disciple which testifieth of these things, and wrote these things: and we know that his testimony is true.’ (John 21:24)? Ah ha! You bring up the single example in all 4 Gospels of this -- which by itself is not particularly convincing. Even there, it is my understanding that this passage may not be in all of the manuscripts, and could totally have been an interpolation. Add on top of that, the d discrepancies between the Gospel of John and the other, doesn't make for a strong case.< I take your point, but I thought you were making a blanket statement, which I then refuted. A single example in 4 gospels, if it covers the gospel it is in, makes one quarter of the gospels. You make it sound worse (from my POV)! Why ‘not convincing’? What would that verse need to say, to make it convincing, in itself? That said, yes, some scholars say Chap 21 is a later addition. But that is not a refutation of my point, which is that the gospels DO make eyewitness claims, and you cannot refute that.

As for contradictions between Paul’s letters and Acts, can you give an example?<< I don't know Acts all that well, but I every time I have looked into it I get things like: https://ehrmanblog.org/does-the-book-of-acts-accurately-portray-the-life-and-teachings-of-paul/ and there are others. The conversion of Paul, his travels, and where he got his information all seem to be different. I also have never seen apologists lean on Acts a whole lot for their case, which I would expect if it had information that would help their case< Thanks for the Ehrman ref. There are not any real contradictions are there?

Ehrman writes ‘the book of Acts states that when Paul went to Athens he left Timothy and Silas behind in Berea (Acts 17:10–15) and did not meet up with them again until after he left Athens and arrived in Corinth (18:5).’ But Acts 17:15 says ‘And they that conducted Paul brought him unto Athens: and receiving a commandment unto Silas and Timotheus for to come to him with all speed, they departed.’ So Ehrman is incorrect.
1 Thess 3:2 ‘And sent Timotheus, our brother, and minister of God, and our fellowlabourer in the gospel of Christ, to establish you’ Timothy went back from Athens to Macedonia and then from there to Corinth.

The narrative in Acts does not describe a 3 years interval when Paul went to Arabia, but is not inconsistent with it. For all we know, there is 3 years between Acts 9:25 and 9:26. Luke does not say ‘immediately’.
Yes Acts puts a different slant on Paul’s attitude to the Mosaic Law than we see in Paul’s letters. But so what? It is quite normal to see this in everyday writing. A biography of a person may put a different slant on that person’s life from his own autobiography, but without any contradictions between them. As Ehrman says, Paul’s attitude to the Law is complicated, that is putting it mildly. Luke simplifies. What good historian does not?

15..

There is no possible world in which 2+2 does not equal 4<< Mathematicians think about the world in a funny way. :-) Here's a somewhat related question. If you were programming a computer to do deduction, of which 2+2=4 is a consequence or if you prefer something a bit more downstream the "internal angles of an isosceles triangle are equal", how would you do it? You'd need some numerical representation of Boolean logic, conventionally taken as 0 for False and 1 for True. If you do this, then the math you write down is entirely consistent with the same math using probability theory and P=0 and P=1. That's all I'm saying.< OK, but you do not address my point. Logical reasoning does not use probability theory, as I keep saying.

In summary, still waiting for your good evidence, while you dance around the issue, making excuses.< What evidence have you asked for? I am not trying to give you evidence for the resurrection, God’s existence, am I? I don’t understand.

You've tried to paint me as unreasonable, suggesting I wouldn't accept any evidence.< I keep asking you what evidence exactly (for God’s existence, miracle last week) you would accept, and you don’t tell me. See above. Because you don’t tell me, I infer that you would not accept any evidence. Prove me wrong, by listing the evidence.

You've tried to imply that priors are just arbitrary, implying that you'd be justified in whatever bias you come in with and that I have no recourse to disagree.< I never said they were arbitrary. I said the opposite, that they are based on background, judgment, preferences … I never said you could not disagree. I said the opposite, that you and I would disagree on our respective priors.

You've tried to undermine my ability to evaluate evidence ahead of time by insisting that history is different than science, implying that we should hold it to a different standard.< I don’t understand this.

You've suggested multiple times that I want absolute proof, implying that I am not reasonable. If I did any of this in a scientific area, I would not be taken seriously. They'd say, "show me the evidence". Instead of trying to guess what evidence I might or might not accept, instead of trying to imply that my standards might be too high, I'd suggest putting some work into providing what you think is your strongest case for God, or the resurrection, or miracles and improving that case with better evidence if it isn't convincing. It's this last step that theists uniformly ignore, and try to excuse, when in the sciences it is encouraged and people do seriously hard work to accomplish.< I have said that I want to know what evidence you would accept for a miracle to be ‘verified’.
'Verified' is your word, so please show me what that means in practice. I am not trying to guess what evidence you would accept, I am asking for it. But you won’t tell me! If you want to get into me setting out the evidence for God, that is a different discussion, and I thought we were winding down! Should we finish on these points first? Then we could start a new thread just on that?

If he acts supernaturally, as would be expected, then it cannot be studied by naturalistic means, with experiments etc.

again, you're conflating two things -- the cause and the effect. I think the word "miracle" does this automatically, which is a bad thing. Let's be clear, there is a claimed effect (e.g. someone was sick and is not anymore, someone had an injury and doesn't have it any more, someone was dead and is now alive) and a claimed cause (e.g. an agent acting supernaturally did it, magic did it, aliens did it, some unknown natural process did it). One can accept that a particular cause (e.g. God) is working in some unknown way, but be able to study the predicted effects which are all natural. No one is saying these miracles are only having supernatural effects. It's been demonstrated again again (regardless of the proposed cause) if the proposed cause produces effects that are indistinguishable from nothing doing it, or adds nothing to the straightforward natural explanation, then it is not likely the cause actually exists. Predictions from psychics, UFO enthusiasts, faith healers, and homeopaths run up against this all the time. Admitting that there is no observation which requires an agent that can violate physical law is an admission that it is not reasonable to believe it.

what it would take for you to accept a miracle and you say it has to be understood in naturalistic terms. Well then, it would not be a miracle, would it?

As I say above, the word miracle mixes the effect and the cause. I would need confirmation of the effect, at a bare minimum to even consider it. There have been many examples which, on the face, seemed to be interesting. In every case that I've looked, the data have been underwhelming so it hasn't gone past that point. Once that hurdle is overcome, then I'd need to see that the effect requires an agent (regardless of method, natural or supernatural) to be compelling. No one has come even close to this.

And suppose he does not interact at all, being either the deist god or the source of all being?

this is possible, but it seems both to be a nearly content-free statement and disingenuous -- if all theists were deists we wouldn't need atheists. :-)

He exists, but likewise cannot be studied! In that case you say we don’t have good reason that God exists.

yup. if he's playing the biggest game of hide-and-seek, then he can't really judge us, right? If he gives me reason itself, and no good reasons to believe in him because the only methods available to me (other then credulousness) he hides from, whose problem is that? surely not mine.

This is your naturalistic mindset again.

nope, that's your excuse for me not buying your snakeoil. This sounds harsh, but your response is the same response psychics and con-men use when you're skeptical of their claim -- "you're just biased against such things!"".

Consider the case of the deistic god, who designed the universe and then let it run without intervening. Suppose he exists, for the sake of argument. He does not interact, and yet the evidence is all around us in the form of the design of the universe. Are you really saying that he left no good reason to believe he exists?

I'd say, if there was good reason to see design in the universe, that might be at least something to point at. We don't even have that! What we do have is, from every direction, are a bunch of stories from around the world supposedly from this-or-that great-being, telling us stuff about how the world is like or how it was made, that now we know to be false. Water flowing before the Earth? Day and night before the sun? Animals made separately from humans? Humanity starting with 2 people? Death completely caused by people's actions? Global flood? Disease caused by spirits?

After all that, we're left with people making excuses for this appalling lack of evidence for this being that, in many of the stories, was obviously visible in his actions but as our ability to confirm these actions disappears.

Be that as it may, the main point is that the creator of all, if he exists, cannot be one of his own creations, cannot be ‘a thing’ in the universe. The sustainer of all cannot also be one of the things sustained.

You have two choices. Either this God interacts with the world, in which case you need to pony up the evidence for the natural effects that can only be explained by this agent. Or, this God doesn't interact, and you need to abandon Christianity because that is not the God of Christianity. There are no other options.

Here are two reviews of a book written 4 years ago by Rupert Shortt, to give an idea of what I am saying.

https://www.theguardian.com/books/2016/mar/24/god-no-thing-rupert-shortt-review-response-new-atheism

(correct me if you don't agree with any of the quotes I use from these articles, in case I'm arguing against a position you're not taking)

From Shortt: The claim made by religious philosophers of a certain kind is not that God can be invoked to plug a gap, but that there must be some fundamental agency or energy which cannot be thought of as conditioned by anything outside itself, if we are to make sense of a universe of interactive patterns of energy being exchanged. Without such a fundamental concept, we are left with energy somehow bootstrapping itself into being.

I find this argument kind of silly, really an argument from incredulity. There are physical models with things appearing out of nothing, uncaused, and in some of those theories (backed by evidence!) entire universes can come into being in such a way. The fact that the author finds this hard to believe is understandable -- quantum mechanics is profoundly unintuitive -- but the way to argue against it is to put up a competing mathematical model, and demonstrate that this model makes better predictions. You can't just complain that you find it personally ridiculous -- that train left 100 years ago.

https://www.telegraph.co.uk/opinion/2016/03/26/god-is-no-thing-but-he-is-in-charge-of-things/

(this one is behind a paywall)

Here also is the Unbelievable podcast discussion on the book.

I'll check it out, thanks! I did watch that entire 3-hour take-down of Calum and Matt. Very interesting.

During the writing of this post, I did listen to this episode. As far as I could tell, Shortt was talking in deepities -- no real content, and the atheist brought this up saying that this wasn't the God most Christians are actually talking about. I kept wanting him to ask "so, is this 'foundation of all reality' an agent that can make choices?" I don't see how he could possibly justify this given his approach. His primary motivation seemed to be his dislike of the idea of infinite regress, but infinities are weird so why should they conform to our intuitions? One has to be really really careful with the math when dealing with infinities -- which physics does all the time, btw.

Depends what you mean by evidence, doesn’t it? God is not a thing that you can put under the microscope.

You seem to like this statement, as if it is a "checkmate", but it really isn't. Where is the evidence of effects that can't be explained with plausible naturalistic explanations and require some kind of agent? That's it.

Do you really imagine that he is such, that you could test his works and existence like that of a (eg) a fruit fly?

You can't have it both ways -- either the effects of God in the world are observable or they are not. If they aren't, then there is not need to posit the God hypothesis and no good reason to believe. If they are, point them out and defend them, and stop making excuses for it. If I said any of these excuses in any other context, people would not take me seriously.

But Psalm 19 says that the heavens declare the glory of God. Have you ever wondered why there is anything at all, why all this existence comes from?

sure! and I'll believe explanations that can actually be explanations, and not "magic did it". I will probably not know the entire answer before I'm gone, but already the story is pretty compelling. We live in an amazing time where we understand more about where we came from than any other time in the human race, and none of it came from our religious stories.

Do we have all the answers? No, but I'd rather be able to say "I don't know" than to make up something or convince myself of something just because it feels good to have an answer.

Yes exactly. And this is true about most statements about the distant past, where by ‘distant’ I mean that the only sources are written, and there are no more available. No eyewitnesses to questions, no more written sources, no DNA evidence, no videos … So differences in probabilities about things like Mark, Jesus’ resurrection cannot be reconciled by probability theory. We are stuck with the evidence we have, and we don’t have a way of reconciling.

It depends on the claim, right? I can imagine some convergence on mundane claims -- say, the city of Jerusalem existed in the first century. Then, we might be able to get convergence with most informed people on somewhat less mundane claims, like Paul wrote the Letter to the Romans. There are some holdouts that all of Paul's letters are not written by him, but by-and-large the evidence can overcome the modestly low prior of that claim. Something with a really low prior, like a resurrection or flying, there could never be any historical evidence to rescue the prior so the conclusions will be nearly entirely dependent on the prior. This is totally natural, and what we'd expect. It is in fact exactly what I've been saying about the limits of historical claims and their evidence.

The atheist alternative hypothesis of grave robbing plus women and Peter experiencing hallucination plus disciples wanting to believe it is just about plausible I think.

Really? How many reports of grave robbing to we have, compared with people being raised from the dead? How about claims of people rising from the dead vs those confirmed to have been? How do you establish that comparison? I don't think the women finding the tomb is one of the "minimal facts", has no external attestation (Paul doesn't even mention it) and has seemingly good literary reasons for inclusion. A grave robbing, plus an expectation to see Jesus, plus a pentacostal-like spiritual experience seems pretty reasonable as one possibility (there are others) -- we have working examples of each of these. None for bodily resurrections, plus many counter-examples, so I don't see the probabilities falling that way.

But there is no way of adjudicating between it and the resurrection hypothesis.

Actually, pointing out that we have observations of the non-resurrection components and none of the resurrection components, and we have counter-examples of the resurrection components, should place resurrection a-priori below the non-resurrection. Now, bringing in the lousy data we have, shouldn't push someone over to the resurrection model unless they are starting with a bias towards it.

Even if we agreed on the Bayes’ factors, we would be stuck on the difference between our priors.

and that's a signal that the data is lousy. If you have a high probability for something, so you believe it, but then you find out that it is highly dependent on priors then you have two choices:

justify your priors (one person's prior is another's posterior)
recognize there actually is no good reason to believe

Yes but you would not concede that any miracle has been ‘verified’, would you?

they haven't! that's the point! same with the psychics, which is why I don't believe them. any new psychic has to do some heavy lifting to get over the previous data. They never seem to want to. Same with the theists -- they never seem to want to put in the hard work, and are amazed when people don't take their claims seriously.

According to your previous statements, for verification you want to identify the mechanism and have proof of the existence of the agent.

Can we stop it already with the word "proof"!!! Please??!! I didn't use that word, so stop trying to say I did. I want evidence, and that is not impossible.

This, for a naturalist, is impossible (as I have already pointed out).

Again, you're making excuses...stop that. Put up the evidence.

So, short of a change in priors (eg me becoming an atheist or you becoming a believer) there would be no convergence.

Seems like you're saying one or more of:

I'm dogmatically biased against any evidence you want to put forward
The evidence that you can put forward is not sufficient to convince anyone
The only good reason to believe is just to believe ahead of time, without evidence

agree? disagree?

I don’t think confidence is the same as P=1, is it?

right, confidence is large P, not P=1.

I would say that P(Trump supporters invaded the Capitol building on 6 Jan 2021) is not quite = 1, and you agree. But we can still claim certainty can’t we?

it's a matter of being pedantic, mostly on my part, but I reserve "certainty" to mean "proof" to mean p=1, so no we can't claim certainty. Colloquially we will use the word "certain" if the probability is so close to 1 that we just don't want to bother with the difference, but I dont do this -- especially in these conversations -- because it is too often misused. Theists love to jump on the possibility that the atheist is demanding too much evidence, and being able to claim that the atheist is setting P=1 or P=0 is a way of doing that.

Try to imagine cases where it is a good thing for someone to break one of the rules above

Yes absolutely we cannot prove them to be desiderata for all reasoning.

Again, can you imagine a case where it is a good thing for someone to break one of those rules?

My point is that you have nothing to say to someone who does not use them, beyond “You are being irrational”.

yup, that's about it. it's like someone saying 2+2=5, I can say you're not following the math rules. if they say, "so what?" can the conversation continue?

but just pointing out the obvious, that you cannot enforce your standard of reasoning onto others (unless you are a worldwide dictator with power to read peoples’ thoughts!)

you're correct, but I think most people -- at least in their own view -- do value being consistent and rational. My entire point really has been to say, if you want to be rational -- almost by definition -- you need to follow the rules set out by Jaynes and others. These lead directly to the rules of probability, so as a corollary, one must follow the rules of probability to be rational.

But if we see above, we see that we seem to agree that convergence is impossible if there are not enough data.

right!

Even if we both have all the info in front of us, we could treat it differently, put different weights on different parts, according to our own background knowledge and preferences.

correct.

It looks like you are (now) agreeing with me.

wow, it does look like this!

There is not an automatic process of convergence.

it can't be automatic, it's a limit dependent on the evidence. if you have zero (literally zero) evidence, then the probabilities are going to be exactly the priors. If you have a little evidence, but not much, then the posteriors will be similar to the priors. Convergence always comes with more data.

A slight wrinkle in this is that when one says "prior", one might not be meaning "prior to all data" but "prior to new data", which is "posterior from the old data". If you're talking about the probability of the sun rising tomorrow, with no data that might be 50-50, but with hundreds of data points it might be 99-1 and that's what you're using for your "prior" today. One has to be careful about that. (see https://bblais.github.io/posts/2014/Sep/22/will-the-sun-rise-tomorrow/)

A ‘study of the origin of our solar system’ is not a law of nature, surely?

no, but it informs laws of nature. The study of the ancient solar system helps to clarify our understanding of gravity and thermodynamics. There is the confirmation of specific events in the past (e.g. the collision which caused the formation of the Moon) and the clarification of physical processes which make predictions for other systems and observations. Just like history, there is the confirmation of specific events in the past (e.g. the rise of the Roman Empire) and the clarification of the sociological/psychological/cultural processes which make predictions for other systems and observations. I don't see these things as particularly different in structure, although they do differ substantially in our ability to interact with the data. Dividing claims into scientific vs historical does not seem to me to be a useful distinction.

You can define it all mathematically, using functions and sets.

No we are not talking about a real pack cards, it is all abstraction.

And probabilities in this simple sense are all unconditional.

Conditional probability and Bayes’ Theorem came later.

I agree with all four of these statements. However, I still don't see it as a useful distinction. You are just saying that prior probabilities predate posterior (both historically and mathematically, it seems).

You are defining science as anything that applies mathematics to the world. But then by definition you are right.

So I'm right! :-) Actually, I prefer to think of it as math is the language of science, so in that way it really is by definition just because it is the language used.

I thought by science you meant modern empirical science, as practiced in the West since c1500. But in your new definition, the hunter-gatherers were using science when they counted how many nuts they have!

Next time you redefine a word, warn me first!

People use inference a lot, but they didn't formalize it until much later. I think of probability as being the formal backdrop of this, much of which is not used explicitly. Hume would have been greatly improved if he could have used probability theory directly. Having the formalism allows one to be much more specific, see where the structure works even in non-quantitative cases, and to see which specific rules one is violating when one deviates from the process.

But the rest of your answer shows we are making progress. Can we be even more specific? What would move YOUR (not anyone else’s) needle for belief in (the Christian) God above 50%?

a) The stars rearranged to spell BRIAN BLAIS BELIEVE IN ME YAHWEH one night and then returned to their original position the following.

Or

b) You personally witnessed a healing miracle, like the one in Times Square where a man’s legs reattach to the body after someone prays in the name of Jesus.

Or …

Please tell me.

As I've said before, I never know what will convince me, or even more-likely-than-not because science is a process, one that I can't foresee the outcome. It could be I am initially moved by some evidence, but then someone comes forward with an alternative explanation which is more compelling. It's only after a lot of that back-and-forth that we arrive at being convinced. However, it should be enough to move the needle in the right direction. ironically, suggestion (a) is so extreme, especially with the lack of any other evidence for God, that my first inclination would be to question my own sanity. (b) is a bit better, but my first inclination would be to suspect trickery. In both cases, I'd need some follow-up that eliminated these options.

So what would start to move the needle for me? My conditions tend to be a bit more modest. What if the specific claims of theists tended to be right more often then wrong? If the texts say there was a global flood, that there is evidence for that independently of the texts. If there was a resurrection to vindicate that Jesus conquered death, how about appearances to more people -- perhaps to the Chinese, who has a printing press at the time? How about the people who say that prayer actually works, can demonstrate that it works above chance and placebo? How about people who claim God can heal you, can show well-documented examples that aren't easy to refute -- amputees, for example. How about a specific claim from a religion that actually makes a specific set of predictions about the origin of life, the universe, biology, something useful? Every time theists make specific claims, and I can investigate them, they fall apart and we're only left with vague claims (i.e. creation story = metaphor) or excuses (i.e. God can't be investigated). The parallels with psychics is remarkable.

Does that help?

I think we agree on this. There is no difference in principle, only difference in the values of probabilities. Exactly, no real difference in principle, just difference in quality and amount of data.

this is the most we've agreed upon from the beginning! :-)

OK, but I think you should have been clearer, when you give words meanings different from the usual. Most people would think ‘prove’ means much the same as ‘establish’, don’t they?

Sometimes people do that, but I try to be more careful, because to do otherwise invites misuse. I've been bitten by that before -- someone using the word "prove" instead of evidence, then I use it that way, then they say that I insist on 100% certainty because "proof" is used in that way in maths.

You demand that an alleged supernatural event be demonstrated as a natural event before you accept it, a contradiction in terms.

No, I demand that an alleged supernatural cause be observed as a natural effect. One can possibly infer the cause even if you don't know its specifics. I am not even sure what a supernatural effect would be, and how we would know it, or care.

We should know better now. Do you disagree?<

No I don’t, taking your point.

amazing! more agreement! yay!

I take your point, but I thought you were making a blanket statement, which I then refuted.

So, going back a month and a half (2020-12-02), I found my original statement about eyewitnesses. Did you refute this?

The claim "people who knew at first hand the details of his life and teaching or people who spoke with those eyewitnesses" is not agreed upon by most historians. Paul was no eyewitness, and the other texts are from anonymous texts. The Gospels (with the exception of one small verse in John) make no claim to being eyewitnesses and that verse is John is also suspect.

Notice I even point out the exception of that one passage in John! I get tired of writing that whole paragraph out, but the bottom line is that there is no good reason for believing the Gospels are written by eyewitnesses or are even informed by eyewitnesses. I also read all of Bauckham, in addition to listening to that entire youtube series.

Why ‘not convincing’? What would that verse need to say, to make it convincing, in itself?

I am no a scholar in ancient literature, but I would expect (at a minimum) something like what other ancient historians did -- name their sources. They even often go so far as to critique those sources, pointing out where they might be mistaken, or not. I understand that the standards in ancient history are far lower than we have come to expect, but even they took some steps to establish dates and sources. We get none of that from the Gospels. even the one passage you site does not actually say the name of the person writing it!

The narrative in Acts does not describe a 3 years interval when Paul went to Arabia, but is not inconsistent with it. For all we know, there is 3 years between Acts 9:25 and 9:26. Luke does not say ‘immediately’.

This kind of reasoning gets us things like https://bblais.github.io/posts/2013/Feb/15/resurrection-linear-regression-and-the-art-of-harmonization/

I imagine that you can harmonize the nativity stories, saying that Luke doesn't mention the flee to Egypt, but doesn't say he didn't go there, and maybe the timeline is adjustable? Acts is not agrees upon by scholars as representing eyewitness testimony, and there seems to be a very clear direction to the differences between Acts and the statements in Paul -- in the direction of making Peter and Paul agree. Again, I'd have to study Acts a lot more for more detail on this, but it doesn't seem as if most theists are willing to have their points dependent on the veracity of Acts.

Logical reasoning does not use probability theory, as I keep saying.

Let's make an analogy, which might help. There are multiple formulations of mechanics in physics. Two of them are Newtonian and Lagrangian. The Lagrangian formulation makes no use of forces, yet is entirely consistent with the Newtonian form and in some cases more convenient. Probability theory is a different way of framing the same problems as traditional Boolean logic, and can generalize to those cases where there is uncertainty. So, it can both be the case that traditional Boolean logic does not use probability or the vocabulary of probability and that Boolean logic is a limiting case of the rules of probability.

What evidence have you asked for?

I am not trying to give you evidence for the resurrection, God’s existence, am I?

At some point, yes. It's not enough to criticize someone's approach to a problem, ask what it means to be considered evidence, to ask what would convince me. At some point, given that the answers have been generally "I'll believe when I see the evidence, do you have any", it is time to present some. I find it much easier to talk about reasoning and beliefs when addressing specific claims.

You've tried to paint me as unreasonable, suggesting I wouldn't accept any evidence.

I keep asking you what evidence exactly (for God’s existence, miracle last week) you would accept, and you don’t tell me.

Because you don’t tell me, I infer that you would not accept any evidence.

Hopefully I've answered this, at least enough to proceed. Not knowing exactly what would convince me is not the same as declaring that nothing will convince me. The proper response, in the case of the former, is to try me out -- show the best evidence, and why it may or may not be convincing.

You've tried to imply that priors are just arbitrary, implying that you'd be justified in whatever bias you come in with and that I have no recourse to disagree.<

I never said they were arbitrary. I said the opposite, that they are based on background, judgment, preferences …

background = data, so we are not dealing with priors. judgment must be based on prior evidence, so we are not sharing the same information -- let's get that info out. preferences are pretty arbitrary. Let's see where the priors are actually coming from, so we can see what is shaping them.

If you want to get into me setting out the evidence for God, that is a different discussion, and I thought we were winding down!

do these discussions ever wind down? :-)

Should we finish on these points first?

Then we could start a new thread just on that?

We can do both-and, depending on time/energy. As for a new thread, you can either find a relevant post I've already done (the search bar works on the site) or I can post something new on a topic you'd like, and then have the thread start from that. either works.

Thanks Brian (guthub seems better than Disqus!)

If he acts supernaturally, as would be expected, then it cannot be studied by naturalistic means, with experiments etc.<< again, you're conflating two things -- the cause and the effect. I think the word "miracle" does this automatically, which is a bad thing. Let's be clear, there is a claimed effect (e.g. someone was sick and is not anymore, someone had an injury and doesn't have it any more, someone was dead and is now alive) and a claimed cause (e.g. an agent acting supernaturally did it, magic did it, aliens did it, some unknown natural process did it). One can accept that a particular cause (e.g. God) is working in some unknown way, but be able to study the predicted effects which are all natural. No one is saying these miracles are only having supernatural effects. It's been demonstrated again again (regardless of the proposed cause) if the proposed cause produces effects that are indistinguishable from nothing doing it, or adds nothing to the straightforward natural explanation, then it is not likely the cause actually exists. Predictions from psychics, UFO enthusiasts, faith healers, and homeopaths run up against this all the time. Admitting that there is no observation which requires an agent that can violate physical law is an admission that it is not reasonable to believe it.<

No I don’t think I am conflating cause and effect. But ‘miracle’ does imply supernatural cause and effect inexplicable in naturalistic terms. Yes we do need to define ‘miracle’, and I suppose there are different definitions. I would say, for our purposes, ‘a supernatural action by a supernatural agent resulting in some effect which would normally be welcomed (healing etc) and which cannot be explained in normal naturalistic terms. But I think you are still blind to the obvious fact that, by definition, a supernatural act cannot be studied naturalistically, ie investigated scientifically. Such (alleged) acts are usually one-off, so they cannot be studied. I have said this a few times to you. I wonder if we are misunderstanding each so completely on this point, that everything we each say goes past the other. Is that so, do you think?

(this exchange moved up to fit with the point above)

You demand that an alleged supernatural event be demonstrated as a natural event before you accept it, a contradiction in terms<< No, I demand that an alleged supernatural cause be observed as a natural effect. One can possibly infer the cause even if you don't know its specifics. I am not even sure what a supernatural effect would be, and how we would know it, or care.< I do not understand you, sorry. Maybe you could illustrate with eg Jesus’ alleged resurrection?

what it would take for you to accept a miracle and you say it has to be understood in naturalistic terms. Well then, it would not be a miracle, would it?<< .As I say above, the word miracle mixes the effect and the cause. I would need confirmation of the effect, at a bare minimum to even consider it. There have been many examples which, on the face, seemed to be interesting. In every case that I've looked, the data have been underwhelming so it hasn't gone past that point. Once that hurdle is overcome, then I'd need to see that the effect requires an agent (regardless of method, natural or supernatural) to be compelling. No one has come even close to this., But you previously said that you would also need the existence of the agent to be ‘established’. You said (about 5 January) … Third, say we've ruled out everything that we could think of (yet) and someone is claiming a specific causal agent (e.g. Christian God), they'd need to establish that the agent actually exists, propose and demonstrate how that agent could interact with the world, etc... otherwise it would contain the same explanatory power as "magic did it".<

What would move YOUR (not anyone else’s) needle for belief in (the Christian) God above 50%?<< a) The stars rearranged to spell BRIAN BLAIS BELIEVE IN ME YAHWEH one night and then returned to their original position the following. Or b) You personally witnessed a healing miracle, like the one in Times Square where a man’s legs reattach to the body after someone prays in the name of Jesus. Or … Please tell me.<< As I've said before, I never know what will convince me, or even more-likely-than-not because science is a process, one that I can't foresee the outcome. It could be I am initially moved by some evidence, but then someone comes forward with an alternative explanation which is more compelling. It's only after a lot of that back-and-forth that we arrive at being convinced. However, it should be enough to move the needle in the right direction. ironically, suggestion (a) is so extreme, especially with the lack of any other evidence for God, that my first inclination would be to question my own sanity. (b) is a bit better, but my first inclination would be to suspect trickery. In both cases, I'd need some follow-up that eliminated these options. So what would start to move the needle for me? My conditions tend to be a bit more modest. What if the specific claims of theists tended to be right more often then wrong? If the texts say there was a global flood, that there is evidence for that independently of the texts. If there was a resurrection to vindicate that Jesus conquered death, how about appearances to more people -- perhaps to the Chinese, who has a printing press at the time? How about the people who say that prayer actually works, can demonstrate that it works above chance and placebo? How about people who claim God can heal you, can show well-documented examples that aren't easy to refute -- amputees, for example. How about a specific claim from a religion that actually makes a specific set of predictions about the origin of life, the universe, biology, something useful? Every time theists make specific claims, and I can investigate them, they fall apart and we're only left with vague claims (i.e. creation story = metaphor) or excuses (i.e. God can't be investigated). The parallels with psychics is remarkable. Does that help?< No it does not, but thanks for trying, even though you don’t answer the question. You are contradicting yourself. On the one hand you think that everyone thinks the same way, treats data the same way. On the other hand you cannot say what data will turn a known prior into a posterior of (say) 0.6. Don’t you see the contradiction? I repeat, what evidence would convince you that the existence of the Christian God ia more probable than not? Please be specific.

And suppose he does not interact at all, being either the deist god or the source of all being?<< this is possible, but it seems both to be a nearly content-free statement and disingenuous -- if all theists were deists we wouldn't need atheists. :-)< Why ‘content-free statement and disingenuous’? I don’t understand. Seems that the proposition ‘The universe was created by an intelligent designer’ has quite a lot of content. And disingenuous, how?

He exists, but likewise cannot be studied! In that case you say we don’t have good reason that God exists.<< yup. if he's playing the biggest game of hide-and-seek, then he can't really judge us, right? If he gives me reason itself, and no good reasons to believe in him because the only methods available to me (other then credulousness) he hides from, whose problem is that? surely not mine.< I'd say, if there was good reason to see design in the universe, that might be at least something to point at. We don't even have that! What we do have is, from every direction, are a bunch of stories from around the world supposedly from this-or-that great-being, telling us stuff about how the world is like or how it was made, that now we know to be false. Water flowing before the Earth? Day and night before the sun? Animals made separately from humans? Humanity starting with 2 people? Death completely caused by people's actions? Global flood? Disease caused by spirits? After all that, we're left with people making excuses for this appalling lack of evidence for this being that, in many of the stories, was obviously visible in his actions but as our ability to confirm these actions disappears.< Him judging us is not the issue. The OT stories are irrelevant here. My point is the fairly obvious one that if we see something with the appearance of design, we naturally infer a designer. You might disagree that the universe shows evidence of design, or that the universe as a whole is not the same type of case as a football found on a beach, but do you accept the general principle of inference of a designer from appearance of a design?

Be that as it may, the main point is that the creator of all, if he exists, cannot be one of his own creations, cannot be ‘a thing’ in the universe. The sustainer of all cannot also be one of the things sustained.<< You have two choices. Either this God interacts with the world, in which case you need to pony up the evidence for the natural effects that can only be explained by this agent. Or, this God doesn't interact, and you need to abandon Christianity because that is not the God of Christianity. There are no other options.<

You ask for evidence for (the Christian?) God.
But I have said that I am not trying, in this discussion of your book, to provide evidence for theism.
That is another discussion. It is also another discussion whether I should abandon my belief in God for lack of evidence. We could talk about that later if you like.

In any case you have not addressed my point. I repeat. The creator of all, if he exists, cannot be one of his own creations, cannot be ‘a thing’ in what he has created.

Yes he might interact in some way, or he might not. But in neither case is he an agent in the universe whose interactions can be studied. If he interacts, his interactions will be one-off. So, for instance, suppose that the Christian God of the universe exists, and that his only action was to raise Jesus from the dead? How could you, now in 2021, possibly study that miracle and ‘verify’ it? There is no way, as I think, you now agree. But he still exists, by supposition! Your naturalism does not allow belief in such a god, even supposing he exists. NB I don’t understand your reference to ‘natural’ effects (see above about the need to explain it).

(sorry about paywall, I will attach the article by Howse)

From Shortt: The claim made by religious philosophers of a certain kind is not that God can be invoked to plug a gap, but that there must be some fundamental agency or energy which cannot be thought of as conditioned by anything outside itself, if we are to make sense of a universe of interactive patterns of energy being exchanged. Without such a fundamental concept, we are left with energy somehow bootstrapping itself into being. I find this argument kind of silly, really an argument from incredulity. There are physical models with things appearing out of nothing, uncaused, and in some of those theories (backed by evidence!) entire universes can come into being in such a way. The fact that the author finds this hard to believe is understandable -- quantum mechanics is profoundly unintuitive -- but the way to argue against it is to put up a competing mathematical model, and demonstrate that this model makes better predictions. You can't just complain that you find it personally ridiculous -- that train left 100 years ago. During the writing of this post, I did listen to this episode. As far as I could tell, Shortt was talking in deepities -- no real content, and the atheist brought this up saying that this wasn't the God most Christians are actually talking about. I kept wanting him to ask "so, is this 'foundation of all reality' an agent that can make choices?" I don't see how he could possibly justify this given his approach. His primary motivation seemed to be his dislike of the idea of infinite regress, but infinities are weird so why should they conform to our intuitions? One has to be really really careful with the math when dealing with infinities -- which physics does all the time, btw.< I don’t think you get the point that Shortt, and many others (I am currently reading David Bentley Hart’s ‘The Experience of God’), are making, as shown by your comment “deepities -- no real content”. The point is about the contingency of existence. Why is there anything at all? Science cannot deal with this, as it deals only with existence. Rowan Williams says this in his review of Shortt ‘He quotes the British scholar Denys Turner to good effect on the fact that “nothing” ought to mean what it says – “no process … no random fluctuations … no explanatory law of emergence”. The problem of origins cannot be defined out of existence, and the highly complex notion of creation by an act that (unlike finite agency) is not triggered or conditioned needs to be argued with in its own terms, not reduced to the mythical picture of a Very Large Person doing something a bit like what we normally do, only bigger.’

You say science models ‘things appearing out of nothing’. That is not the ‘nothing’ of non-existence, because that ‘nothing’ is some kind of quantum vacuum, aubject to various laws. If it were not subject to those laws, you would not get things appearing out of it! Shortt (and many others) are talking of the ‘nothing’ of nonexistence.

Depends what you mean by evidence, doesn’t it? God is not a thing that you can put under the microscope. Do you really imagine that he is such, that you could test his works and existence like that of a (eg) a fruit fly?<< You seem to like this statement, as if it is a "checkmate", but it really isn't. Where is the evidence of effects that can't be explained with plausible naturalistic explanations and require some kind of agent? That's it. You can't have it both ways -- either the effects of God in the world are observable or they are not. If they aren't, then there is not need to posit the God hypothesis and no good reason to believe. If they are, point them out and defend them, and stop making excuses for it. If I said any of these excuses in any other context, people would not take me seriously.< Your reply here shows your naturalistic blinkers. I think you are conceding (also see above) that if God did exist you would not be able to see evidence for him. See above about supposing that God did only one supernatural act 2000 years ago. See elsewhere about your refusal to accept ANY supernatural acts (you would first need proof of the existence of the agents …). Your dilemma is apparently quite powerful, but on reflection it is a false dilemma. You accuse me of conflating cause and effect but you are doing the same. We may ‘observe’ an apparent miracle eg the Times Square example. That might well be ‘observing the effects of God’. But you will deny that it is an effect of God because you have said that you want proof of the existence of God first (I have quoted you several times on this). So, if God really did exist, and did the miracle, you would be unable to ‘observe his effects’. You would have observed something strange, yes, but you would deny that it was the effect of God because it was a one-off, and you have not seen proof of the existence of God enough to satisfy you. So, some people may be able to observe ‘the effects of God’, but others not. Moreover, as discussed above. What about the deist god? We see the universe as apparently designed, and some conclude a designer, but others do not. The dilemma, superficially quite a powerful one, rather dissolves when it is looked at closely.

But Psalm 19 says that the heavens declare the glory of God. Have you ever wondered why there is anything at all, why all this existence comes from?<< sure! and I'll believe explanations that can actually be explanations, and not "magic did it". I will probably not know the entire answer before I'm gone, but already the story is pretty compelling. We live in an amazing time where we understand more about where we came from than any other time in the human race, and none of it came from our religious stories. Do we have all the answers? No, but I'd rather be able to say "I don't know" than to make up something or convince myself of something just because it feels good to have an answer. Yes exactly. And this is true about most statements about the distant past, where by ‘distant’ I mean that the only sources are written, and there are no more available. No eyewitnesses to questions, no more written sources, no DNA evidence, no videos …< Do you really think science will ever be able to find out why there is anything at all? (See above as well)

So differences in probabilities about things like Mark, Jesus’ resurrection cannot be reconciled by probability theory. We are stuck with the evidence we have, and we don’t have a way of reconciling.<< It depends on the claim, right? I can imagine some convergence on mundane claims -- say, the city of Jerusalem existed in the first century. Then, we might be able to get convergence with most informed people on somewhat less mundane claims, like Paul wrote the Letter to the Romans. There are some holdouts that all of Paul's letters are not written by him, but by-and-large the evidence can overcome the modestly low prior of that claim. Something with a really low prior, like a resurrection or flying, there could never be any historical evidence to rescue the prior so the conclusions will be nearly entirely dependent on the prior. This is totally natural, and what we'd expect. It is in fact exactly what I've been saying about the limits of historical claims and their evidence.< Yes I think we agree. For the resurrection, you and I could agree about the significance of all the evidence, thus having the same Bayes’ factors, but the difference between our priors would result in different posteriors – yours very small and mine > 0.5.

The atheist alternative hypothesis of grave robbing plus women and Peter experiencing hallucination plus disciples wanting to believe it is just about plausible I think.<< Really? How many reports of grave robbing to we have, compared with people being raised from the dead? How about claims of people rising from the dead vs those confirmed to have been? How do you establish that comparison? I don't think the women finding the tomb is one of the "minimal facts", has no external attestation (Paul doesn't even mention it) and has seemingly good literary reasons for inclusion. A grave robbing, plus an expectation to see Jesus, plus a pentecostal-like spiritual experience seems pretty reasonable as one possibility (there are others) -- we have working examples of each of these. None for bodily resurrections, plus many counter-examples, so I don't see the probabilities falling that way. But there is no way of adjudicating between it and the resurrection hypothesis. Actually, pointing out that we have observations of the non-resurrection components and none of the resurrection components, and we have counter-examples of the resurrection components, should place resurrection a-priori below the non-resurrection. Now, bringing in the lousy data we have, shouldn't push someone over to the resurrection model unless they are starting with a bias towards it.< I don’t think you understood me. I was conceding that the alternative hypothesis to resurrection was plausible.

Even if we agreed on the Bayes’ factors, we would be stuck on the difference between our priors.< and that's a signal that the data is lousy. If you have a high probability for something, so you believe it, but then you find out that it is highly dependent on priors then you have two choices:

justify your priors (one person's prior is another's posterior)

recognize there actually is no good reason to believe< Depends on the prior, and also the subjective importance of belief about the topic. There is not much good evidence for lots of things in early history, but it does not bother most people because they do not care a fig about eg Babylonian society in 1500BC and see no need to have a belief about it! So there is a third option

Not have a belief about it Incidentally I think many non-religious people in the West today are in category 3 about religion, in particular things like the resurrection of Jesus. They have heard about it, but do not care enough about it to have a view either way. It does not matter enough to them to have a belief. But interestingly for people like you it DOES matter to have a belief, and it matters enough for you to write a book about religious belief. I wonder why this is so. Do you have a view why some non-religious have such a strong anti-theistic program, and care enough to write books etc. but most do not?

Yes but you would not concede that any miracle has been ‘verified’, would you? they haven't! that's the point! same with the psychics, which is why I don't believe them. any new psychic has to do some heavy lifting to get over the previous data. They never seem to want to. Same with the theists -- they never seem to want to put in the hard work, and are amazed when people don't take their claims seriously. According to your previous statements, for verification you want to identify the mechanism and have proof of the existence of the agent. Can we stop it already with the word "proof"!!! Please??!! I didn't use that word, so stop trying to say I did. I want evidence, and that is not impossible.< OK, sorry. You actually said they'd need to establish that the agent actually exists, propose and demonstrate how that agent could interact with the world, etc... < So you require that it be established, ‘not proved’, that the agent exists. How do you do this? What evidence would do the job? See above.

This, for a naturalist, is impossible (as I have already pointed out).<< Again, you're making excuses...stop that. Put up the evidence.< You are changing the topic, as pointed out above. I am asking you what evidence you would need for the existence of God to be established, in your eyes. We are not talking about me trying to prove God to you.

So, short of a change in priors (eg me becoming an atheist or you becoming a believer) there would be no convergence.<< Seems like you're saying one or more of:

I'm dogmatically biased against any evidence you want to put forward

The evidence that you can put forward is not sufficient to convince anyone

The only good reason to believe is just to believe ahead of time, without evidence agree? disagree?< Disagree. You have a rather one-dimensional view of belief. It is not just considering belief for P in light of evidence for P conditional on nothing else. In fact our beliefs are a rather complex interconnected web, based around what some people call ‘worldviews’.
The worldviews of the theist and naturalist are profoundly different, as we are discovering. My belief in God comes from many things in my life, and a view of the historical evidence for Jesus’ resurrection is only one little thing in that pile. I would suggest that your atheism and naturalism also comes from many things in your life, maybe including upbringing, and that you did not yourself consider the historical evidence for the resurrection with a blank slate. Perhaps you did not start studying it until you were a convinced atheist?

but just pointing out the obvious, that you cannot enforce your standard of reasoning onto others (unless you are a worldwide dictator with power to read peoples’ thoughts!)<< you're correct, but I think most people -- at least in their own view -- do value being consistent and rational. My entire point really has been to say, if you want to be rational -- almost by definition -- you need to follow the rules set out by Jaynes and others. These lead directly to the rules of probability, so as a corollary, one must follow the rules of probability to be rational.< Yes, perhaps since the Enlightenment a majority of people in the West value being rational. But I think you ignore how precarious a hold rationality has! Only since the Enlightenment … Only in the West … Only a majority, and perhaps just a small majority in these days of fake news and post truth. I wonder how many Trump supporters value rationality? I think more than half. What proportion, do you think? Fundamentalists of all persuasions would value the word of their scriptures over rationality – that is a lot of people. People educated in authoritarian states value obedience to the Party or the Leader more than rationality. Also, for the many millions in the world who do not have enough to eat, for instance, they would understandably give up ‘rationality’ for a good meal. After all, rationality does not put food on the table. So I hope you see how your high view of rationality as a desideratum is rather limited to people like you: well-educated, fairly well-off, liberal inhabitant of a Western democracy. Do you disagree?

There is not an automatic process of convergence<< it can't be automatic, it's a limit dependent on the evidence. if you have zero (literally zero) evidence, then the probabilities are going to be exactly the priors. If you have a little evidence, but not much, then the posteriors will be similar to the priors. Convergence always comes with more data.< I disagree. Again, you assume that everyone must think the same way, treat data the same way. I know you would like to think this true, but see just above about your belief that everyone is or wants to be ‘rational’.

A ‘study of the origin of our solar system’ is not a law of nature, surely?<< no, but it informs laws of nature. The study of the ancient solar system helps to clarify our understanding of gravity and thermodynamics. There is the confirmation of specific events in the past (e.g. the collision which caused the formation of the Moon) and the clarification of physical processes which make predictions for other systems and observations. Just like history, there is the confirmation of specific events in the past (e.g. the rise of the Roman Empire) and the clarification of the sociological/psychological/cultural processes which make predictions for other systems and observations. I don't see these things as particularly different in structure, although they do differ substantially in our ability to interact with the data. Dividing claims into scientific vs historical does not seem to me to be a useful distinction.< No there is not much difference in the logic between past events in the solar system and past events in human history. They are both past events. What I was trying to point out that past events are a different kind of thing from laws of nature, with different kinds of truth conditions.

You can define it all mathematically, using functions and sets. No we are not talking about a real pack cards, it is all abstraction. And probabilities in this simple sense are all unconditional. Conditional probability and Bayes’ Theorem came later.<< I agree with all four of these statements. However, I still don't see it as a useful distinction. You are just saying that prior probabilities predate posterior (both historically and mathematically, it seems).< No I am not. I am saying that, in the development of probability, unconditional probability predate conditional probability and Bayes’ Them.

You are defining science as anything that applies mathematics to the world. But then by definition you are right. I thought by science you meant modern empirical science, as practiced in the West since c1500. But in your new definition, the hunter-gatherers were using science when they counted how many nuts they have! Next time you redefine a word, warn me first!<< So I'm right! :-) Actually, I prefer to think of it as math is the language of science, so in that way it really is by definition just because it is the language used. People use inference a lot, but they didn't formalize it until much later. I think of probability as being the formal backdrop of this, much of which is not used explicitly. Hume would have been greatly improved if he could have used probability theory directly. Having the formalism allows one to be much more specific, see where the structure works even in non-quantitative cases, and to see which specific rules one is violating when one deviates from the process.< If you want to define ‘science’ as any practical activity involving maths, fine. Just don’t expect others to understand you, unless you warn them first!

So, going back a month and a half (2020-12-02), I found my original statement about eyewitnesses. Did you refute this? The claim "people who knew at first hand the details of his life and teaching or people who spoke with those eyewitnesses" is not agreed upon by most historians. Paul was no eyewitness, and the other texts are from anonymous texts. The Gospels (with the exception of one small verse in John) make no claim to being eyewitnesses and that verse is John is also suspect. Notice I even point out the exception of that one passage in John! I get tired of writing that whole paragraph out, but the bottom line is that there is no good reason for believing the Gospels are written by eyewitnesses or are even informed by eyewitnesses. I also read all of Bauckham, in addition to listening to that entire youtube series.< OK I get you. But saying ‘one small (?) verse in John’ is misleading isn’t it? Read at face value it covers one quarter of the gospels. And why call it small? You read Bauckham, all the way to the end? Well done!

Why ‘not convincing’? What would that verse need to say, to make it convincing, in itself?<< I am not a scholar in ancient literature, but I would expect (at a minimum) something like what other ancient historians did -- name their sources. They even often go so far as to critique those sources, pointing out where they might be mistaken, or not. I understand that the standards in ancient history are far lower than we have come to expect, but even they took some steps to establish dates and sources. We get none of that from the Gospels. even the one passage you site (sic) does not actually say the name of the person writing it!< That is not an answer to my question. You said that John 21:24 was not convincing and I asked you what would that verse need to say, to make it convincing, in itself.

Commenting on what you do say, I don’t think that the authors of Matthew, Mark and John were seeing themselves as historians. I think that they were writing as best they could memoirs of Jesus for the believers, and certainly that they had no idea that their works would be examined by sceptical non-believers 2000 years later. I myself am sceptical of some of the things in Matthew eg the flight to Egypt, the manner of Judas’ death, the resurrection of the saints. By contrast I think the author of Luke-Acts did aim to write something like history. Note his attempts to be exact about dates, names and titles. Note also his prefaces, where he makes it clear that he is trying to write an accurate record, based of course on sources (written and eyewitness) plus his own eyewitness account from Acts 16 onwards. It is true that Luke does not name his sources, and many people wish that he did and wonder why he didn’t. But still, Luke-Acts is prima facie an attempt at historical writing.

The narrative in Acts does not describe a 3 years interval when Paul went to Arabia, but is not inconsistent with it. For all we know, there is 3 years between Acts 9:25 and 9:26. Luke does not say ‘immediately’.<<s This kind of reasoning gets us things like https://bblais.github.io/posts/2013/Feb/15/resurrection-linear-regression-and-the-art-of-harmonization/ I imagine that you can harmonize the nativity stories, saying that Luke doesn't mention the flee to Egypt, but doesn't say he didn't go there, and maybe the timeline is adjustable? Acts is not agrees upon by scholars as representing eyewitness testimony, and there seems to be a very clear direction to the differences between Acts and the statements in Paul -- in the direction of making Peter and Paul agree. Again, I'd have to study Acts a lot more for more detail on this, but it doesn't seem as if most theists are willing to have their points dependent on the veracity of Acts.< No I don’t believe a flight to Egypt can fit into Luke’s narrative Luke 2:2 says ‘And when the days of her purification according to the law of Moses were accomplished, they brought him to Jerusalem, to present him to the Lord’ Purification is 30 days max, so I say no time for a quick trip to Egypt. I have debated this on a few forums, and seen the various fixes proposed. The most popular is to say that Matt 2 is about 1-2 years later (so it took the Magi a really long time to get there!). Paul’s 3 years in Arabia is different. Luke, like everyone writing history, compresses to leave out things he does not want to mention.

Logical reasoning does not use probability theory, as I keep saying<< Let's make an analogy, which might help. There are multiple formulations of mechanics in physics. Two of them are Newtonian and Lagrangian. The Lagrangian formulation makes no use of forces, yet is entirely consistent with the Newtonian form and in some cases more convenient. Probability theory is a different way of framing the same problems as traditional Boolean logic, and can generalize to those cases where there is uncertainty. So, it can both be the case that traditional Boolean logic does not use probability or the vocabulary of probability and that Boolean logic is a limiting case of the rules of probability< OK but that is not what you were saying about logic/maths vs probability. You said that the former is a branch, subset, of the latter.

I am not trying to give you evidence for the resurrection, God’s existence, am I?<< At some point, yes. It's not enough to criticize someone's approach to a problem, ask what it means to be considered evidence, to ask what would convince me. At some point, given that the answers have been generally "I'll believe when I see the evidence, do you have any", it is time to present some. I find it much easier to talk about reasoning and beliefs when addressing specific claims< We have never talked about me trying to prove God to you. I never said that I could, and I don’t think anything I said would satisfy you. I never even implied that I could.

You've tried to imply that priors are just arbitrary, implying that you'd be justified in whatever bias you come in with and that I have no recourse to disagree.<< I never said they were arbitrary. I said the opposite, that they are based on background, judgment, preferences …<< background = data, so we are not dealing with priors. judgment must be based on prior evidence, so we are not sharing the same information -- let's get that info out. preferences are pretty arbitrary. Let's see where the priors are actually coming from, so we can see what is shaping them.< No, background does not equal data. Your background is all your whole life up to now, including genetic factors, upbringing, experiences as well as acquired knowledge. It is manifestly hard if not impossible to see where priors come from. If it were easy, psychology would be a sought-after discipline. I raise this above. Can you detail the ‘data’ in your life that have resulted in you being an atheist / naturalist? I can point to my upbringing and three seminal experiences, but not ‘data’ as such. I mean, could you ever describe a conversion experience as just ‘data’?

do these discussions ever wind down? :-)< I guess we will have to stop at some point, but I am finding this continually interesting, even though at some points we go round in circles.

Should we finish on these points first? Then we could start a new thread just on that?<< We can do both-and, depending on time/energy. As for a new thread, you can either find a relevant post I've already done (the search bar works on the site) or I can post something new on a topic you'd like, and then have the thread start from that. either works< Gosh I don’t know. This has been fun, but I am not sure I can handle doing more than one of these long posts every couple of days. Like you I am a teacher (I teach higher maths in school), so it is busy in term (teaching everything remotely of course). But on the existence of God, as I said above, I don’t think anything I could say would satisfy you. Furthermore I don’t think that many people actually come to faith from historical study – although J. Warner Wallace’s Cold Case Christianity is an exception. Let us see how this one goes in the next week or so? Maybe soon we will decide we have said everything we want to say on these points?

Regards David

bblais / bblais.github.io

posts/2019/Jul/15/a-measure-of-faith-probability-in-religious-thought/ #2

A Measure of Faith - Probability in Religious Thought · bblais on the web