The fact of the matter, as far as I know, and as I thought anyone would realize is that Bayes’ theorem is a theorem which follows from certain axioms. Its application to any real world situation depends upon how precisely the parameters and values of our theoretical reconstruction of a real world approximate reality. At this stage, however, I find it difficult to see how the heavily feared ‘subjectivity’ can be avoided. Simply put, plug in different values into the theorem and you’ll get a different answer. How does one decide which value to plug in?
^^^exactly.
But subjectivity is part of our normal Humanity no one can get out of subjectivity. What Bayes theorem does is force us to acknowledge our subjectivity it's not willy-nilly arbitrary to set probability. Probability is based on evidence and our background knowledge. In Bayes theorem if you think something is likely that means over 50% Bayes theorem formalizes our argument and makes us put numbers to it we already think in mathematical terms Bayes theorem just formalizes that. If you are pretty certain of a claim that means you think it's higher than 50% probability in fact higher than 80% probability even possibly 90% probability. If someone doesn't agree with your subjective probability then they can discuss the evidence for why they don't think very likely means 80% or 90% or whatever percent they think it is and we carry on the discussion. Subjectivity against Bayes theorem is a red herring. If we say subjectivity is a reason not to use Bayes theorem all reasoning is subjective that suggesting we don't come to any reasons or conclusions about anything which is silly.
Dr CamNC4Me
"Dr. Peterson and his Callithumpian cabal of BYU idiots have been marginalized by their own inevitable irrelevancy defending a fraud."
Seems like Bayesian reasoning is just spitting out what your own judgment has already come to. I just don't see where it objectively demonstrates anything the critical method doesn't already provide. The allure of its appearance towards more certainty seems obvious.
mikwut
All communication relies, to a noticeable extent on evoking knowledge that we cannot tell, all our knowledge of mental processes, like feelings or conscious intellectual activities, is based on a knowledge which we cannot tell. -Michael Polanyi
"Why are you afraid, have you still no faith?" Mark 4:40
Analytics wrote: I'll agree with Philo and his link on this one; I can't think of a clear-headed way of examining the historicity of Jesus without evoking Bayes'.
What if you are a frequentist? I think Bayseian probability can both serve as a way to formalize the inferred reasoning we use and to generate insights that aren't immediately intuitive. But to insist that an empirical question can only be addressed through Bayes Theorem suggests a hardcore Bayesian position that is far from self-evident.
Regarding this thread, you definitely see people around the field of philosophy of religion and issues related to apologetics make dense Bayesian arguments where the Bayesian notation only serves to obfuscate what should otherwise be an accessible chain of thought. And, unfortunately, often-times those are terrible arguments where the use of Bayesian probability is just creating a thin patina of sophistication through use of fancy notation. I say this as someone who is at least conversant enough with Bayesian analysis to be able to read the material.
This is usually happens on the side of religious apologists. I've read more awful "God probably exists as demonstrated by this Bayesian jibber-jab" arguments than I care to recall. I have a sneaking suspicion the same would be going on with Carrier here. I'd have to read more of him on this subject.
If you are a frequentist, then you are a simpleton, lol.
As I said earlier, to the extent an argument is valid, it is implicitly Bayesian. Ideally, being more explicit about the Bayesian framework should make your argument more clear—not less so. The really big advantages to an explicit Bayesian approach is that it forces you to consider the conditional probabilities in their own right. That Goodacre debate is a good example. Goodacre's arguments were repeatedly of the form, "it wouldn't make sense for Paul to be a mystic, because the people he was talking to all personally knew the historical Jesus." Hopefully it's clear why that is fallacious. If he was endeavoring to be more explicitely Bayesian, hopefully he wouldn't have fallen into this unconvincing line of reasoning.
It’s relatively easy to agree that only Homo sapiens can speak about things that don’t really exist, and believe six impossible things before breakfast. You could never convince a monkey to give you a banana by promising him limitless bananas after death in monkey heaven.
Analytics wrote: He has a PhD in history from Colombia, so he needs to be taken seriously.
I haven't followed Carrier much since he got his Ph.D. I'm skeptical he's changed all that much. I know him still as the guy who gradually rose the ranks in the atheist online community at iidb. He was good about some things, bad about others, and had a little too much of a self-promoting, off-putting personality. This comment is amusing as all get out to me.
It'd be like Tal Bachmann getting a Ph.D., then hearing someone hear assert that we now have to take him seriously because, hey, he's got a Ph.D.
All I'm saying is that he has earned the right for his arguments to be seriously considered rather than dismissed out of hand.
It’s relatively easy to agree that only Homo sapiens can speak about things that don’t really exist, and believe six impossible things before breakfast. You could never convince a monkey to give you a banana by promising him limitless bananas after death in monkey heaven.
Seems like Bayesian reasoning is just spitting out what your own judgment has already come to. I just don't see where it objectively demonstrates anything the critical method doesn't already provide. The allure of its appearance towards more certainty seems obvious.
mikwut
What, specifically, is "the critical method?"
Anyway, I'm about to run out of the country for a week and a few days, and won't be on the Internet while I'm gone. I will be reading that Carrier book though. I shall return and report.
Cheers!
It’s relatively easy to agree that only Homo sapiens can speak about things that don’t really exist, and believe six impossible things before breakfast. You could never convince a monkey to give you a banana by promising him limitless bananas after death in monkey heaven.
The traditional collection of methodologies and strategies historians use to understand the past. The criteria or tools historians use to understand the past. Historians simply speaking use evidential based reasoning, logic, critical evaluation of the known evidence, attempt to avoid bias, and weigh the evidence in the light most favorable towards the best explanation. In Jesus studies critical methodologies of form, redaction, and source criticism are used to best understand the text from the authors own time and culture based on the evidence available to us. Simply put it is how historians are taught to think and do their work. It is this method or criteria that Carrier and Philo believe has failed because we have a multiplicity of Jesus constructions by applying these criteria to the known evidence and therefore the emphasis on Bayes reasoning is emphasized to clarify this multiplicity.
As I said earlier, to the extent an argument is valid, it is implicitly Bayesian. Ideally, being more explicit about the Bayesian framework should make your argument more clear—not less so.
Exactly and precisely! We couldn't agree more, which is why I have been arguing that it is just a different language put to probabilities (maybe it does clarify for some, great) but it isn't, indeed, can't be, radically different than historical weighing of the evidence and interpreting that towards the best explanation because those are the same arguments inputed into a bayesian formula. This is why I argued earlier to Philo that biologists use a similar criteria that historians use when they reconstruct our biological past. If we apply Bayesian reasoning something is radically wrong if we end up with creationism rather than a clarification of evolutionary history. That is my main argument in this thread. if Bayesian argumentation is going to make the argument more clear then how on earth respecting Jesus scholarship does it do the opposite? One of the clearest things we know from the traditional historical method is Jesus existed, but somehow when carrier uses Bayes theorem that clarity is turned on its head. Either Bayesian reasoning is malarkey, the basic evidential weighing and reasoning of the traditional historical method is malarkey - or, Carrier is just inputing into Bayesian reasoning the radical fringe arguments of mythicism and using Bayesian reasoning to raise his credibility.
have a great trip!
mikwut
All communication relies, to a noticeable extent on evoking knowledge that we cannot tell, all our knowledge of mental processes, like feelings or conscious intellectual activities, is based on a knowledge which we cannot tell. -Michael Polanyi
"Why are you afraid, have you still no faith?" Mark 4:40
richardMdBorn wrote:I was one of three guests on a two hour Chicago radio program last year. The host raised the issue of Drake's equation about the probability of advanced civilzations existing outside of the earth in the Milky Way. Another guest, who is a NASA lunar geologist in Houston, scoffed at DE since he doesn't think any of the terms can be estimated. How do you know if the probability of term X is one in ten versus 1 in 10^100 in DE. In the same way, how do you assess the strength of the gospels' evidence for Jesus' existence versus Roman references to him from the second century versus Josephus' two references to him (acknowledging the problems with one of the two references). Providing an equation where all of the terms cannot be estimated may make you feel sophisticated, but it's a worthless exercise in my opinion.
Excellent point. Drake's Equation is useful for stimulating speculation but not for arriving at conclusions.
I agree with you. Both of the formulas can be useful if they help you list all of the considerations which are pro and con to a given position. Thus, one must consider with regard to the historical Jesus: 1. Literary sources New Testament Jewish Roman
2. Archaeological sources Pilate stone Jerusalem digs
3. Other messianic stories from the time
Listing items helps ensure that you don't leaving out anything important. But any probabilities who are associated with them are useless in my opinion.
richardMdBorn wrote:I've heard this argument before but I don't think it works. And if proto gospels were floating around, they wouldn't necessarily need to reinvent the wheel. Note that Luke, the author of Luke-Acts was Paul's companion (see the change from third to first person in Acts 16). One can make a strong argument that Luke wrote both Luke and Acts, with Luke clearly written first. Acts ends with Paul still alive. It's a relatively short book. If Paul had been executed before the completion of Acts, surely his death would have been included. Thus, Acts was written before 65 and Luke before that. Ancient sources claim that Mark was based on the recollections of Peter. Thus, the attempt to split the gospels from the Pauline-Petrine writings does not work as well as you think it does. I note that you have a very late date for Mark (c 80).
Acts could have been written decades after Paul's death, and then ended the way it did because that is when the person writing it ran out of paper or died himself.
I was going to say something similar to Mikwut in regard to advancing our empirical knowledge about the world, but let's not condemn the theorem for its misuses. If it does tell us how we form judgments, then that in itself is huge. Perhaps its value isn't so much in what it predicts, but what it can tell us about our assumptions. For instance, if someone judges the Book of Mormon to be true, then Bayes Theorem shouldn't tell us why the judgment is wrong, but how that judgment makes sense in regard to the rest of the person's world.
The way I've always seen Bayes Theorem used -- and this might just be due to my limited exposure to the world -- is similar to Ockham's Razor, someone is trying to either prove or disprove something about religion that's controversial. Apologists appear to love Bayes Theorem. If you can get really good at it, you can load the dice and perhaps outwit a solid skeptic who has good senses but not a statistics background.
I would definitely side with Reverend Kishkumen, but I'd take it farther, and say that Bayes Theorem is totally useless for any field for the same reason Ockham's razor is. For the amusement of Analytics, I'll point out that my philosophy of science is rooted directly in the Efficient Market Hypothesis (that's from economics, for others). In that view, Bayesian calculations and appeals to simplicity find an analogy to metrics such as a Price/Earning ratio. Suppose the price of a stock is X and the P/E ratio is Y, and according to P/E, X is way to high. In one view of things, the market must be wrong since the price is not justified by solid company fundamentals, but in the EMH view, all the influential market players know what a P/E is, and so the stock price must be driven by something the market knows but that the metrics aren't capturing.
What does that mean for science? It means that if science believes X, but Bayes Theorem tells us Y, then we're going to stick with the scientific consensus. But suppose science is wrong, and that in a particular field no one had ever heard of Bayes Theorem, and by use of Bayes Theorem, a scientist makes a startling discovery that turns the field on its head. In that case, scientists in that field will now use Bayes Theorem like crazy, until the theorem no longer gives an advantage to any one scientist over another. Once scientists are properly informed, and a consensus stands that is add odds with Bayes, now what? Can't happen?
As an analogy, there is a similar heuristic often called out by skeptics: falsifiability. Falsifiability is most often used against religion, but it's also been used as an argument against string theory by a minority of scientists who reject string theory. So is string theory wrong if it isn't falsifiable? Surely, every living string theorist knows what "falsifiability" is, even if in a few cases the knowledge is implicit only: it's been considered, and yet, the answer remains at odds with this heuristic. Whatever can usefully be extracted from falsifiability, Bayes, Ockham's razor, or any algorithmic scientific method has already been extracted by the professionals of any given field and reflected in the consensus of that field. In the EMH analogy, that's kind of like saying all available information is reflected in the stock price.
To be clear, while useful to a degree for instruction, all these heuristics at once become pretty well useless for weighing in on any scientific debate (from the EMH view). If the consensus is historical Jesus and Bayes theorem disagrees, then academics know something the theorem isn't capturing. If the consensus is not a historical Jesus and Bayes agrees, then it's a happy accident and doesn't follow Bayes, therefore no Jesus.
Lou Midgley 08/20/2020: "...meat wad," and "cockroach" are pithy descriptions of human beings used by gemli? They were not fashioned by Professor Peterson.
LM 11/23/2018: one can explain away the soul of human beings...as...a Meat Unit, to use Professor Peterson's clever derogatory description of gemli's ideology.
Gadianton wrote:I was going to say something similar to Mikwut in regard to advancing our empirical knowledge about the world, but let's not condemn the theorem for its misuses. If it does tell us how we form judgments, then that in itself is huge. Perhaps its value isn't so much in what it predicts, but what it can tell us about our assumptions. For instance, if someone judges the Book of Mormon to be true, then Bayes Theorem shouldn't tell us why the judgment is wrong, but how that judgment makes sense in regard to the rest of the person's world.
The way I've always seen Bayes Theorem used -- and this might just be due to my limited exposure to the world -- is similar to Ockham's Razor, someone is trying to either prove or disprove something about religion that's controversial. Apologists appear to love Bayes Theorem. If you can get really good at it, you can load the dice and perhaps outwit a solid skeptic who has good senses but not a statistics background.
I would definitely side with Reverend Kishkumen, but I'd take it farther, and say that Bayes Theorem is totally useless for any field for the same reason Ockham's razor is. For the amusement of Analytics, I'll point out that my philosophy of science is rooted directly in the Efficient Market Hypothesis (that's from economics, for others). In that view, Bayesian calculations and appeals to simplicity find an analogy to metrics such as a Price/Earning ratio. Suppose the price of a stock is X and the P/E ratio is Y, and according to P/E, X is way to high. In one view of things, the market must be wrong since the price is not justified by solid company fundamentals, but in the EMH view, all the influential market players know what a P/E is, and so the stock price must be driven by something the market knows but that the metrics aren't capturing.
What does that mean for science? It means that if science believes X, but Bayes Theorem tells us Y, then we're going to stick with the scientific consensus. But suppose science is wrong, and that in a particular field no one had ever heard of Bayes Theorem, and by use of Bayes Theorem, a scientist makes a startling discovery that turns the field on its head. In that case, scientists in that field will now use Bayes Theorem like crazy, until the theorem no longer gives an advantage to any one scientist over another. Once scientists are properly informed, and a consensus stands that is add odds with Bayes, now what? Can't happen?
As an analogy, there is a similar heuristic often called out by skeptics: falsifiability. Falsifiability is most often used against religion, but it's also been used as an argument against string theory by a minority of scientists who reject string theory. So is string theory wrong if it isn't falsifiable? Surely, every living string theorist knows what "falsifiability" is, even if in a few cases the knowledge is implicit only: it's been considered, and yet, the answer remains at odds with this heuristic. Whatever can usefully be extracted from falsifiability, Bayes, Ockham's razor, or any algorithmic scientific method has already been extracted by the professionals of any given field and reflected in the consensus of that field. In the EMH analogy, that's kind of like saying all available information is reflected in the stock price.
To be clear, while useful to a degree for instruction, all these heuristics at once become pretty well useless for weighing in on any scientific debate (from the EMH view). If the consensus is historical Jesus and Bayes theorem disagrees, then academics know something the theorem isn't capturing. If the consensus is not a historical Jesus and Bayes agrees, then it's a happy accident and doesn't follow Bayes, therefore no Jesus.
Dean, you nailed it in your first paragraphs. Bayes has very limited applicability unless we can define all elements under scrutiny in metrical terms.
I would be qualified in support of String Theory. The criticisms of it as a general unifying theory continue and are substantial. Many physicists are willing to recognize it in describing quantum gravity well but not particle physics. More tests, proposals, revisions are under way.
