On Saturday, February 9 (2013), I will be debating the proposition “Is the Christian Faith Reasonable?” with Dr. David Marshall at the University of Alabama in Huntsville. The whole event will run from 6pm to 8pm, doors open at 5:30pm, and Dr. Marshall and I will be selling and signing our books in the lobby afterward. The debate will be held in the Chan Auditorium. For more details see the calendar page for UAH Non-Theists, the co-sponsors of the event with the campus group Ratio-Christi.
Get Dr. Carrier’s Latest!
About the Author
Richard Carrier is the author of many books and numerous articles online and in print. His avid readers span the world from Hong Kong to Poland. With a Ph.D. in ancient history from Columbia University, he specializes in the modern philosophy of naturalism and humanism, and the origins of Christianity and the intellectual history of Greece and Rome, with particular expertise in ancient philosophy, science and technology. He is also a noted defender of scientific and moral realism, Bayesian reasoning, and historical methods.
Support Dr. Carrier
Books by Dr. Carrier
Explore C.H.R.E.S.T.U.S.
Road trip!
Where do I buy tickets?
I think it’s open to the public, but you might want to check the linked site and see for sure. I assume they don’t expect to run out of seating. But who knows if that’s a sound assumption or not. Might just have to be sure you are there at doors-open.
David Marshall’s name pops up in so many reviews or comments to reviews about Amazon books, that it would be instructive to give us a rundown on this writer, his resume,his strengths and his weaknesses (I was able to spot only weaknesses), his stand, and what degree of credibility we can attribute to him (beyond my too subjective evaluation).
Too often he seems to write a review or a comment to a review without having read the book, only to expatiate on his own credo.
I, for one, had come to the conclusion that his Amazon writings were not only tendentious, but empty of real analysis or authentic critique.
You likely know about this writer much more than we do, and would be willing to draw a portrait of him as a man and as a critic.
If anyone here has avidly read his reviews, they would be better positioned to do that. I am only familiar with some of his behavior in this regard (and that mainly only in his treatment of me). I have certain notions, but proving them would be a job of work. I’d rather just focus my time and energies on the coming debate. But I welcome others to talk about him here. As long as its civil and analytical and constructive (which means it can be negative and critical, just in a soundly argued way).
The link I provided to the main sponsor contains a bio for Marshall. But for more, John Loftus has discussed Marshall’s reviews of his own work in ways that can be illuminating.
If anyone else has any other good links like that, do feel free to post them here (you can use A HREF tags in comments).
Look at it this way. The Atheist “A” will blend right in. Which is good, because in this area the only thing they love more than God is trying to pronounce the word “God”. Don’t even think of calling him Yahweh either or they’ll wonder why you’re asking about their way.
Alabama, come for the moonshine, stay to find your teeth the next morning.
(I’m about 5 and a half hours further south than Huntsville, so I’m not speaking as some damn Yankee)
5 and a half hours? Do you live in the Gulf, or Gainesville?
Wow, into the lion’s den!
His blog is Christ the Tao, and I will concede that I was rather amused by his recent entries proposing that Muslims and Secular Humanists ought to celebrate Christmas.
It’s almost like a Muslim stating that Christians ought to fast on Ramadan because Jesus Christ was one of Mohammed’s predecessor prophets, and thus a proto-Muslim.
DM’s “fulfillment” theory seems based on selective quotation, quotes out of context, and gross misunderstandings, complete with ignoring loads of counterevidence. If he tried hard enough, I’m sure that he could “prove” that Xianity is the “fulfillment” of both Marxism and Randism.
DM also claims that Xianity produced modern science, and he has also claimed that the “Gospel of Christ” has done more to liberate women than anything else. Geez, I didn’t know that there were all these Xian priestesses over the centuries.
Theists usually lost debates when faced with New Atheism adversaries, promoting classic philosophical atheism. However now, with the advent of Atheism+ they finally have the chance to win – they just need to forget about absoulte stuff from ultimate source and just compare – is Christianity more reasonable than the cult called Atheism+? The answer might be as well YES. Debate won.
“If anyone else has any other good links like that, do feel free to post them here”
I don’t know if you are interested, but David Marshall has a long history of arguing with Arizona Atheist, who now blogs at Skeptic Ink. Even if you don’t agree with everything he says, i think there are a lot of information that could be useful.
http://arizonaatheist.blogspot.it/search/label/David%20Marshall
Will there be a video or at least an audio available??
I believe it will end up online for free viewing at some point. But that can’t always be guaranteed.
Actually, I have a long history of ignoring Arizona Atheist. Although talking with a certain philosopher this evening about an upcoming project we’re cooperating on, I realized that history should probably have been even more complete than it has been. But Richard should feel free to borrow anything he likes from Ken.
Roo is out to lunch. Of the 300 odd books I’ve reviewed on Amazon, I’ve only failed to read the bulk of the text in the case of the two or three worst stinkeroos, which were so bad one could hardly get past the title page without one’s fingers firmly on one’s nose. And no, my “credo” does not deflect me from giving strong praise to books by atheists, Buddhists, or Hindus, if they write well, or “thumbs down” to books by Christians and Republicans, if they write poorly. In other words, I review with integrity and because I love good books: anyone who can’t see that, is a fool. Put all my reviews together in one group, and they’d have some 8000 “helps” votes, which would make me one of the top Amazon reviewers of serious books.
As for Ipetrich’s more interesting comments, those subjects might have to wait for a different debate — looks like I’ll be doing a few more this year, including a couple on the impact of Christianity. But maybe I’ll blog on his / her comments about Fulfillment Theology, which is probably the best (if sketchily developed here) approach a skeptic could take to FT arguments (I dealt with it in the first chapter of my doctoral thesis), if there’s a lull in the storm.
You are probably going to debate from the standpoint of a historian, thus using some analogies from ancient times, but for me the major reasons why Christianity is not reasonable are reltaed to the standards we are used to in the modern world.
Consider the description of how the world began, how it works etc. contained in Christian literature. Where are the equations? Where are nubmers, constants? Where is the raw data? In the modern world without them any paper of on how the world works or how it began is soundly rejected. Of course there are popular positions, like “Short History of Time”, but what they do is they refer the main body of work that has hordes of equations, numbers and is based on explicitly shown raw data. If the Bible is such popular literature, where is the real deal hidden?
Then there’s the data size problem. We live in the age of Big Data, where petabytes of data are gathered, processed. Google saves something like 3TB per day. Large Hadron Collider produces something like from 15PB to 25PB of data per year. The Large Synoptic Survey Telescope (2016) is expected to produce 30TB of sky images per night of observation, with the goal to create 150PB database. Notice this Big Data is for analysis and for discovery. As of now, despite Petabytes of probing no God was found in this data. How many probes, of what spatial/temporal/frequency etc. ranges and what accuracy would be enough to say “There is no God”? Getting few tiny probes is apparently enough to claim that food is safe, so why should’t that many probes be enough for conclusion that there is no God?
Also how about the moral laws do they fulfill the requirements of the modern world? Today the major actors are institutions, eg. corporations. There is no system of moral laws on how corporations should behave included in the Christianity. A God’s justice is supposed to having humans torutred for eternity for acts like stealing, while there’s no punishment for corporations for doing similar stuff for which humans are supposed to be tortured forever – what would be the equivalent btw? And it’s clear that you can’t just apply rules for humans to corporations. Corporations are expected to kill other corporations, eat them, and desire sexy parts of other corporations. A typical humanist stupidity would be to blame some humans that work for corporations for eventual evils done there. That’s stupid, as a corporations can make decisions using comptuers, or through a distributed decision structure where each processing element (a human or a computer) does so little part of processing that it is below any threshold of responsibility.
Yet another critical aspects is caring for the environment. It is both a moral and a legal issue, currently firmy established in secular morality and secular laws. Christianity looks ignorant there. Also quite naive “Give a man a fish, and you feed him for a day; show him how to catch fish, and” … he’ll overfish and thus cause fish extinction.
I’m told the debate will be live streamed here.
I should have checked this comments section before the debate, but I didn’t and I missed the streaming. In the near future, will you post your thoughts on how well/not well the debate unfolded?
I felt a little bad for Marshall. The debate did not go well for him at all. It was a clear win for the negative. The video should become available eventually, and when it does I’ll blog about it.
It was actually an informative debate, and Marshall did not resort to any underhanded tactics (well, one red herring, but a mild one, and I called it out in the debate), and he avoided most of the sillier arguments of fundamentalists. Perhaps that technically tied his hands: with those bogus arguments one can shotgun their way to victory, hoping I’ll run out of time to rebut them all, even though they are so easily refuted (and would always be if the clock never ran out); Marshall perhaps was aware of that, and his not doing it suggests a measure of honesty in defense of his beliefs.
Nice job in this debate Richard. I’m glad you had occasion to address the illusion of veracity given to Craig’s positions simply because of his technical wins within the time constraints of oral debates. And although Mr. Marshall seems like a decent guy, this debate was like a barrage of softball fallacies on the affirmative side. Anyway, kudos on a good performance!
Is there any audio/video of the debate posted online anywhere?
There is or will be, supposedly. I’m still awaiting word. Keep your eye on this thread.
I just tracked down the podcast on this debate, and I enjoyed listening!
I have a fairly basic question about the debate tactic when it comes to the origins of the universe especially, especially given how WLC loves Kalam. It seems to me that this is bait that, if taken, will suck you down the rabbit hole. I know the temptation is to say “there is a plausible explanation” — but let’s be honest, the plausible hypotheses are (by my understanding) still quite speculative, and far from what we would call a well-supported scientific theory. This tends to degenerate quickly, and 98% or more of the audience doesn’t have a clue. If they don’t have a clue, then they don’t have the opportunity to experience cognitive dissonance, which to me, is how I define winning a debate.
I think it better to simply say that nobody knows, there’s lots of speculation, active area of research, yada yada, but for right now, nobody really knows. I suspect one day we will know, and when we do know with great certainty how the universe began, we know that it happened by purely natural processes. Then, I suspect that the the christian will claim that this purely natural process is the process by which god used to produce the universe, and that god is actually present in some other gap in human knowledge. We know this because it’s been the pattern for the last 500 years.
Same with abiogenesis. And fine tuning. I really don’t have a problem saying “we don’t know”, and if the opponent does claim that they know, then they’re obligated to demonstrate with scientifically verifiable evidence, the same standard that we hold scientists to.
So my question is: why does Richard (and others) engage in the details? To me, the arguments don’t come across as persuasive. I would take a cue from Hitchens and turn it into an opportunity to go on the offensive and say something like the following: When we don’t know, we say we don’t know. Admitting you don’t know is much better than pretending to know because you have an ancient story about people that lived to be 900 years old and had a conversation with a talking snake. Now who are you going to trust: scientists that say we don’t know, or the book with the talking snake that got everything else about cosmology wrong.
Can you direct us to which podcast you mean?
That only works when you have time to explain why this rebuts the argument actually being made at the time. Debates by definition rarely afford us much time to do that. Thus, oral debates demand different tactics than real-world debates, which can only occur in print.
I don’t recall this argument being much of an issue in the debate you are referencing (my opponent dropped it pretty quickly; creationism does not entail Christianity anyway, so it was already a moot argument for him, as I pointed out). But usually when it features, their argument is “God exists because there is no other way to explain x than God.” The only valid rebuttal to that argument is to present counter-evidence refuting it. What is counter-evidence refuting it? A way to explain x without God. It does not work to say “you don’t really know what explains x, no one does,” since that is the fallacy of gainsaying (just denying what they said without proving it).
The argument from ignorance only becomes valid when it is properly articulated, which takes a lot more time than you are likely ever to have in a debate. Thus, for example, it’s my lead argument in my book, Sense and Goodness without God (III.3, p. 71). But only because I can back it up with an extensive discussion of epistemology and warranted belief (which precedes that) and an extensive discussion of the attempts by Christians to dodge that argument by insisting their explanations for x are nevertheless more probable than ours (which follows that).
Which gets me to reason number two that this tactic doesn’t work in oral debates, which is that as soon as you attempt the argument by gainsaying (Them: “God exists because there is no other way to explain x than God”; You: “You don’t really know that”), they counter-rebut with what is actually a valid, even if unsound, argument from greater probability (Them: “we can explain x with God; no one else can explain x; if we all have only one working explanation and no others, we ought to conclude that that one explanation we know is more probable than others we don’t know”), which then devolves into a debate over just whether God actually explains x (which will run out the clock and all your at-bats before you get anywhere to your advantage).
If, instead, you propose alternative explanations of x, they will answer with “ours is more probable than those,” and then you respond by showing that that statement is false. The audience then at least goes away realizing that “God” is neither the only nor even necessarily the best explanation for x. Which eliminates x as a reason to believe in God. Which is the objective of the debate (for the atheist, that is).
In fact, cognitive dissonance is only ever created by showing that the structure of the universe is actually more likely if not made by God (the whole argument of my chapter on the design argument in The End of Christianity, which I can deploy, and have deployed, very successfully in abbreviated forms in oral debates like this one). Because that makes God not fit the evidence. And that hurts. And yet to do this, we don’t even have to say we know what is the cause of x; all we have to point out (as I did) is that whatever it was, it clearly isn’t likely to have been a God.
The basic failure of Beginning of Universe type arguments for the existence of gods is that they don’t answer the question a debate which is typically a present-tense one – do gods exist NOW. Even proving the existance of deities 14 billion years ago doesn’t mean they persisted to the modern era. Apologists deny their gods the right to die, or commit suicide or whatever other way to cease to exist is. This is usually done by a false assesment that “such being has to be timeless” (whatever this means, though – timeless, spaceless, immaterial sounds like nothing, not like a god). Which doesn’t follow the argument – for the creation of the universe plus abiogenesis a hypothetical creator has to be present just for first 10+ billion of years, not forever.
Typical arguments from the beginning of the universe also have a big hole in the sense that they are trying to deny the possibility of gods arising after the beginning of the universe. How can they do it if they allow the existance of supernatural in general, but totally fail to mention the mechanisms of it’s generation, its coming into existance. By the way – can gods arise by evolution with natural selection? Or be created by intelligent designers of some previous civilizations?
Another subject is a general problem of dealing with entities that are much more powerful than us. If we alllow the possibility that there are begins of extreme power that are trying to influence us, then they would be able to cheating us, by providing us perfectly falsified evidence for any subject, that we with our limited powers won’t be able to recognize as false. As such the assumption of existance of extremaly powerful beings results in the impossibility to make conclusions for us. Thus an assumption or conclusion of existance of powerful gods means it impossible to prove anything, including the existance of gods.
Concerning arguments from fine tuning – you really need to indicate theists that this is a perfect example of thinking backwards! It’s obvious that the non-biological part of the universe was first, that in it’s enormousness it is not influenced by the biological part at all, while it influences biological part heavily (eg. the asteoid that killed dinosaurs) and that it is the biological part that is trying hard to adopt to conditions that are created by non-biological part. With methods of adaptation that include dying off of unfortunates and being replaced by more adapted species.
And the “we don’t know” part is a great opportunity to show how we should discover new things, what kind of work is done to do this, and that relying on faith and waiting for revelations doesn’t lead to meaningful discoveries.
Richard,
I can’t find the link. Maybe I searched for it, gave up, and listened to another debate; it was late. It’s not critical to this question, though. I’ve seen it several times in several debates, and I’ve seen you very intentionally going down that (my opinion) rabbit hole.
Thanks; that does help me understand the strategy. I agree completely with this:
Yes, perfect, and I agree; you use this to great effect! And yes, you do have to provide specific evidence to support that, but I think there is concrete evidence (which you cite effectively) that is not speculative. E.g., the amount of time it took for multicellular life to evolve. I would add the existence of parasites, viruses, carnivores, killer asteroids, etc., things that would be immoral to have created.
But where I disagree is here:
Ah, I see! This is where I think the formal rules of debate diverge from what I actually care about. I am perfectly happy admitting that there may be some phenomenon X that is best explained by God. Honestly, the goddidit explanation does explain a heck of a lot of things, which is why it is so intuitively attractive. I also realize that the argument to the best explanation does then follow to support the theist position.
But all is not lost! If the best explanation has a probability of being correct of 0.0000001, then the best explanation should be rejected as almost certainly wrong, and not tentatively accepted! I understand the rules of debate, but here is where I would be happy to technically lose points in a debate in order to put forward a more persuasive argument. So this may be a difference of strategy and style. But again, I think one can simply rebut by highlighting the abysmal track record of the god hypothesis, and paint the opponent as arguing that “maybe this time the god hypothesis will be vindicated” — and agree that it might, but given the record the probability is small, and that one is justified in believing it only after it’s shown to be actually true, and not merely asserted. In contrast, natural explanations have an exemplary track record of being correct 100% of the time.
I also fully appreciate the problem of limited at-bats in a formal debate, and I am no expert on this. In that kind of limited debate, I take the philosophy of the best defense is a good offense. Again, a style thing, you’re damn good at playing defense, but I prefer to go on the offensive and press the attack if given limited time. Going offensive has greater potential for resonating with someone and creating cognitive dissonance. (My debates are all small and informal, so I’m not speaking as anyone with authority.)
Again, in my assessment, nobody is going home from a debate caring if M-theory is correct or adequately disproven, so I stay away from it. (I’m a geek; I care, but I realize I’m a small minority.) But if a few people go home thinking “maybe god really is just a human-made construct used to explain the unknown”, then I would feel like I’ve won. Because I honestly think that’s the truth, and I honestly have no idea if string theory is true or not.
I’m not implicitly arguing that my approach is better; I’m really trying to understand why you intentionally choose to engage, and partly explain why I have chosen not to. Thanks!
That’s technically incorrect (or at least, it’s skipping past the actual point). In Bayesian terms, the only way to get P(T|E) = 0.0000001 is if the priors and consequents get you there. And there’s the rub. You can’t just jump immediately to asserting P(T|E) = 0.0000001. You have to actually demonstrate that P(T|E) = 0.0000001. And that’s where we end up in the debate cycles I described.
For example, there are two ways to get P(T|E) = 0.0000001 (many more, but all others are just variations on these two):
(1) You can have P(E|T) = P(E|~T) but P(~T) = 9,999,999 x P(~T). Translation: the evidence is equally likely whether God exists or not, but God has a prior probability that is vanishingly small relative to alternative explanations of the same evidence. But the former assumes we have explanations that make x as likely as God does (thus, “we don’t know” can’t get you there) and the latter has to be proved somehow (and that’s not easy to validly do without any proposed alternative explanation).
For example, “we don’t know what caused x, but based on centuries of accumulated data we do know that what usually causes things like x are godless objects and forces–so far, in fact, without exception; therefore, even if we don’t know what caused x, the cause of x is still almost ten million times more likely to be some ~T unknown to us than any T we can presently conceive.” But that argument is hard to pull off, since if no cause of x is even conceivable other than T, the reference class fails (x is now no longer like any other thing, therefore what’s been true of other things isn’t true of x). Thus, some ~T still has to be at least conceivable for the “argument from low prior” to apply (we don’t need to know that that specific ~T is true, but we do need to know that there is at least one ~T that is ten million times more likely than T based on past precedents in our search for the causes of things; therefore at least one ~T must be at the very minimum conceivable, or conceivable at least as much as T is).
And in any case, one has to argue for this low prior for T. Since prior probability is relative (P(~T) is always the converse of P(T) and their priors must sum to 1), this means we have to be able to argue for P(~T) being extremely high, not just for P(T) being extremely low (otherwise, if we have no argument either way, then so far as we know, and thus so far as we must admit, P(T) = P(~T), which is what we’ll get to next). How do you argue that P(~T) is extremely high, when you can’t even conceive of what P(~T) is? I’m not saying that can’t be done, but you have to recognize it is perplexingly hard to do, and perhaps far too hard to do in a time-crunch situation like an oral debate. Thus, we have to at least be able to state some possible ~T’s that can be argued to have a higher prior probability than T.
(2) You can have P(T) = P(~T) but P(E|~T) = 10,000,000 x P(E|T). Translation: T and ~T are equally likely so far as we know (this describes a state of total ignorance regarding the cause of E, such that it could just as likely be T as ~T; if it is not just as likely to be T as ~T, then we are not in a state of total ignorance: we must know something about ~T, per above), but we have some evidence E (which must be part of the x to be explained) that is extremely unlikely on T but completely expected on ~T (or else very much more expected on ~T than on T, even if very improbable either way). This is the argument I use in TEC, per my previous remarks.
Although I also there argue that P(~T) >= 3 x P(T), that’s a long way from P(~T) >= 9,999,999 x P(T), and in any case I show why P(~T) must be >= 3 x P(T), in a manner that requires me to at least have some idea what ~T might be. The main weight of the argument is born by showing that P(E|~T) >= 1000s x P(E|T), all things considered (or P(E|~T) = 6 x P(E|T), when only one thing is considered at a time and the others ignored, but that’s just an argument a fortiori). But the only way to show that E is expected on ~T is to have some idea what ~T is. We don’t need to know exactly what it is, any more than a theist can claim to know exactly everything about T (God). But some ~T at least has to be as conceivable as T in explaining E. That’s the only way to show that E is probable on ~T but not probable on T, or that E is more probable on ~T than on T.
Thus, arguments from ignorance don’t really work in this exchange. We always have to have some idea of what ~T could be. And usually when we use an argument from ignorance, we’re sneaking that in the back door, by convincing ourselves there is surely some ~T that does as good a job at explaining E as T does. A theist can of course gerrymander his T to get T to explain E as well as any alternative, but doing that greatly reduces P(T). However, explaining that is hard work in an oral debate where the audience doesn’t know Bayes’ Theorem or really anything about the logic of evidence. The simplest way to communicate the trick’s invalidity is to reference Occham’s Razor and “argument from convenience” (just conveniently inventing exactly the right God for every item to be explained), which if a theist gets to do that, so does a naturalist (so no gain can be made that way), but again explaining that is time consuming. [Look up “gerrymander” in the index to Proving History for more on that]
P.S. I’m now reminded of an apt example, where I make the very argument you have in mind, vis-a-vis the cause of qualia, in TEC, p. 300. Note what I have to state there for the argument to work, in particular that the present conceivability of explanations of qualia on ~T is on par with that for T (i.e. God does not explain qualia any less mysteriously than physicalism does, much less other nontheisms like Taoism).
In other words, we can conceive of there being a physicalist explanation for qualia, based on past precedents of confirmed causal explanation in cognitive science, but we cannot conceive of what that explanation precisely will be; however, the theist is in the same position, since they can’t make predictions regarding qualia or explain why God has, created, or manages them any more than physicalism presently can. Thus the evidence is a wash: it argues neither for nor against theism.
This is not the case for cosmology, where we can easily conceive of numerous explanations for all the evidence, including why there is something rather than nothing, and so on. The theist is actually in a worse position here, since countless proposed cosmological theories successfully predict scores of data points (without inserting any god into their models) whereas almost no useful predictions in respect to the data can be made from any creationist theory (except, notably, predictions that are falsified, as I point out in TEC).
Ah, yes, I think I see the disagreement. Let me jump to it.
To start, how does one argue for the small P(T)? First, I must note that I am not actually arguing for a small P(T). More precisely, I am using a low prior probability that God is the agent of X. I really don’t have to say anything about T. My beliefs are perfectly consistent that a God that exists and doesn’t actually do anything, for example. So I’m not exactly sure how to capture this distinction with your notation, but I think it’s critical. I’ll look at this some more. I’m not entirely sure what your distinction is between E and X, for example.
Remember the theist argument here is that P(T) is large because P(T caused E) is large, where E is the universe coming into existence. It’s that T caused E part that is critical. This isn’t expressed as a conditional; it’s the direct causation that is used to infer existence. I think you’re trying to make the problem too difficult.
If I can demonstrate that P(T caused E) is actually rather small, then I have succeeded. THAT is what I use the principle of historical analogy to do. I think you’re folding some of this into a small P(T), but I don’t think that’s necessary. I don’t need to demonstrate that god is highly improbable a priori. All I need to do is show that it is highly improbable that God is the causal agent for an observed phenomenon for which there is no confirmed natural explanation. P(E|T) doesn’t capture that, nor does a small P(T). I’m not sure how to express that with your notation, so I won’t run the risk of trying and adding to the confusion.
Second, I don’t understand what you mean by “can’t conceive”. Earlier, under case (1), you write “…since if no cause of x is even conceivable other than T…” I don’t know exactly what you mean by the term conceivable. I’m able to conceive of an explanation, but it’s very probably wrong. Where does that get us? I don’t see much distinction between the two cases: A) X is conceivable, but all current concepts are probably wrong, and B) We presently have no concepts. Both are transient, and both have numerous historic parallels, and many of those involve theistic claims of supernatural agency. So far, all of the historical theistic agency claims have been wrong. That’s significant.
Last, I’m confused by the statement from what I quoted above: “this means we have to be able to argue for P(~T) being extremely high, not just for P(T) being extremely low…” Um, wow, I’m really confused! P(T) + P(~T) = 1, yes? So arguing P(~T) is nearly 1 implies that P(T) is nearly zero… right?
Then you are in scenario (2), where P(T) = P(~T). My analysis for that then follows.
That amounts to the same thing, when we are talking about causal hypotheses for X. To wit…
Indeed. And I include that possibility in my demarcation in my chapter on this in The End of Christianity as well. But this, call it T*, entails a low P(E|T) for any design argument. Thus you can’t argue for T* from design. Because T* is inert, it doesn’t produce evidence (much less of design). It’s therefore moot to this debate. Unless instead of atheism you want to defend deism, but then you have to actually argue for T*, which requires you to argue against ~T*. Which entangles you in the same web all over again, since the theist is claiming only T explains X; saying T is false because T* is true is not a useful rebuttal unless you can actually prove what you say, which entails a long digression in defense of atypical theism, which IMO is a waste of an atheist’s time. You need to find a committed deist to do that.
X is the generic description of the evidence (“the universe”) or a biased selection of its properties, while E is all the actual evidence encompassed by X (all the actual contents and attributes of the universe, not just the ones they want to focus on). This is a common theistic fallacy, to argue things like “T caused the universe” but fail to actually include everything that is actually entailed by “the universe” (like arguing “T committed the murder” but leaving out of “the murder” the fact that it occurred a thousand miles away from where T was at the time). Jeffery Jay Lowder I think wrote a good article on the constant use of this fallacy by theists (or perhaps it was someone else, I can’t recall). I’m looking for that now. I’ve asked Jeff for help finding it again.
It always has to be a conditional. This is required by the logic of evidence. P(T|E) is the conditional “the probability that T if E” and Bayes’ Theorem proves that that probability is always the mathematical product of two other conditionals, P(E|T), “the probability that E if T,” and P(E|~T), “the probability that E if ~T,” plus the prior probability of T (the probability of T independent of E), which always entails knowing the prior probability of ~T.
That amounts to the same thing I am talking about. You still have to do this by scenario (1) or scenario (2), or some variation thereon.
And you can’t do it by defending T* unless you actually prove T* is more probable than T, per above. Which entails a huge time-wasting digression, per above.
That depends on what you mean by “very probably wrong.”
If you mean the sum of the posterior probabilities of all conceivable ~Ts is very small, then you are conceding the theist’s argument that the sum of the posterior probabilities of all conceivable Ts is very large.
If you mean the sum of the prior probabilities of all conceivable ~Ts is very small, then you are conceding the theist’s argument that the sum of the prior probabilities of all conceivable Ts is very large. Then the only way to refute T is via a variant of scenario (2), wherein the evidence is even less probable on T than P(~T) is less than P(T). That would have to be some really good evidence.
If you mean the posterior or prior probability of any one ~T is very small but the sum of the priors or posteriors for all ~T is very high, then P(~T) or P(~T|E) is very high, even when every conceivable ~T is very improbable. I explain this in my discussion of a similar scenario in respect to resurrection apologetics in my Reply to Davis:
I think this is the scenario we are in vis-a-vis cosmology (except there the godless options are more numerous than just four).
Correct. My point is that these entail the same task, i.e. because your last statement above is true, you can’t argue that P(T) is extremely low *unless you can also argue* that P(~T) is extremely high. In other words, if you *can’t* argue that P(~T) is extremely high, then you can’t argue that P(T) is extremely low.
Richard,
Thank you for your thoughtful and thorough response. I think that we are still somehow talking past each other. I will try one last time. For context, we’re discussing debate strategy to counter Kalam. I understand what you’re saying, and it’s all factually true, but I think it misses one important aspect of as (I think) properly applied to this scenario.
First, let me clarify: I don’t actually believe in a deistic god. I just don’t care to distinguish between no god and a god that doesn’t care. Second, I understand Bayes theorem and its application here, but for simplicity I’ll leave off the “given the evidence” part in my formulation below.
Rather than try to refine the notation, let me use the Davis example as a way to keep it simple and make my point. Let’s say there are five competing positive hypotheses, labeled A, B, C, D, and H, where H is the goddidit hypothesis. We can assign probabilities that each is correct. Let P(A) = P(B) = P(C) = P(D) = 1%. Let P(H) = 6%. The point here is that they do not add up to 100%.
There’s the remaining 90%, which we can attribute to the null hypothesis, or the unknown alternative, or whatever term you like. This represents both the lack of supporting evidence for all proposed hypotheses, as well as the likelihood that the correct hypothesis is still not yet known and thus undefined. Let’s call it the null hypothesis N, and then let P(N) = 1 – (sum of all others) = 90%.
Note that this is a different problem than, say, trying to decipher whether a signal is a zero or one, or the best explanation of the empty tomb claims. Both of those are “noisy” systems with uncertainty, and you have to make the best estimate among defined alternatives. In those instances, you only care about the relative probabilities, choose the one with the maximum likelihood, and move on. But science doesn’t work that way. The right answer may not be known, and may not have even been conceived of yet. Science doesn’t work by picking the most probable; it works on convincing skeptics beyond unreasonable doubt. You will hear scientists say that some theory X is the leading theory at the moment (i.e., P(X) is relatively large compared to alternatives), but we have a long way to go before we really know anything (i.e., P(X) is small in an absolute sense).
I think that is exactly the condition that best represents the situation where we are at when it comes to the universe beginning to exist. I’m being generous to H here, and maybe overly critical of A-D, but you see my point. The theist is arguing that P(H) > P(A), and perhaps even P(H) > P(A,B,C,D), therefore T. Or they may be arguing that P(A,B,C,D) is small, therefore T. Both are unsound. I think the best counter-argument is not that P(A,B,C,D) is actually not-so-small after all, but that P(~H) is actually quite large because P(N)=90%. Further, in the past when we’ve been in situations in which P(N) = 90%, the correct hypothesis was a naturalistic one 100% of the time, so there’s good reason to be hyper distrustful of H this time.
Furthermore, I’m being very generous to H here, because there really is no evidence for H. The arguments for H are: a philosophical argument (“out of nothing, nothing comes”), and an argument from ignorance. The philosophical argument is a re-statement of the argument from ignorance: “We have never seen something come from nothing, and therefore it can’t.” This also is not sound. We have never seen something come from “nothing” because we have never seen “nothing”! How can we say anything about what “nothing” can or can’t do? We don’t even know that “nothing” can even exist. This exposes the theistic position as over-reaching.
What the theist actually needs to do is:
1. Have a robust theory of “nothing”
2. Experimentally validate the theory of “nothing” extensively and convincingly
3. Demonstrate that it is impossible for something to come from “nothing”
4. Demonstrate (not merely assert) that before the universe came into existence, there was “nothing”
5. Demonstrate with positive evidence that the universe came from “nothing” uncaused
If the theist can do all that, then, well, the theist would have a good argument! They could then present a strong case for deism. But the theist is a very, very long way from making his case.
I’m getting a little off track, so let me close. Bottom line, I think you’re missing the null hypothesis N in your reasoning, and I think that properly including that makes the argument much stronger. Big fan, got a couple of your books, enjoying them, and appreciate being able to interact via your blog.
–Dan
True, if we’re talking *only* Kalam, then I might revise some of my remarks, I don’t know; I was assuming the context of a fine tuning or similar argument.
Kalam can be refuted in 60 seconds by simply pointing out there is no logical connection between it’s penultimate and ultimate conclusion (there is no syllogistic link between “there was a first cause” and “the first cause was a personal being,” that step is always a big hand waving trick) and therefore it’s useless. It can at best prove a quantum wave function started the time-space continuum, if we combine its penultimate premise with the best established facts of science.
There are also defects in all its other premises, too, of course. Big Bang theory does not support a first point in time, thus Kalam’s second premise cannot be established; and the evidence supporting the first premise is not categorically analogous to time itself, thus that everything that begins to exist in time has a cause affords no evidence that time itself began to exist or had a cause, in fact the evidence of everything in time is that causes precede what begins in time, but it is logically impossible for anything, much less a cause, to precede the beginning of time so as to have caused it, so the first premise is not only not established, it is demonstrably false–in fact it is necessarily false, since there is one thing for which it cannot, in any possible universe, be true: time.
(Likewise, arguments against infinite series are mathematically illiterate.)
So I consider the Kalam argument to be an outright joke. I never take it seriously. Hence I wasn’t imagining that’s what you had in mind.
That said, though, you still aren’t getting the Bayesian reasoning quite right:
I’m assuming you are talking about prior probability, otherwise it is unclear how you think you would get any of the probabilities you list. Prior probability is a relative probability, not an absolute probability. It’s the probability of A relative to B, etc. But the breakdown you outline suggests you think it’s the other way around.
For example, if there are only two logical possibilities for how you got rich, you earned it or you won the lottery (set aside for now that there are other possibilities in reality; I’m just trying to make the point simple), the prior probability of H : “you won the lottery” is not the probability of winning a lottery. It’s the baseline probability of a rich person having gotten rich by winning the lottery. In the scenario constructed, if half of all rich people earned it and half got rich by winning the lottery (which logically can be the case even if the odds against winning a lottery are a million to one, and even if only a dozen people out of millions have ever won a lottery, it all depends on how many rich people there are), then the prior probability of H is 0.5. Even when the probability of winning a lottery is a million to one or the probability of being a rich lottery winner is a million to one. As long as E = “you are rich.”
Once that’s understood, it becomes unclear how you could ever know that, for example, the prior probabilities of all known cosmological theories sum to 10% leaving 90% for all unknown theories. How would you know that all known theories are ten times less likely than any unknown theory? I’m not saying that’s impossible, I’m saying it doesn’t seem likely given the present state of cosmological science and philosophy.
Moreover, all that matters is the prior probability of the sum of all “known” cosmological theories relative to the prior probability of the sum of all “known” creation theories. If cosmological theories are ten times more likely than creation theories, it literally does not matter what the sum of the prior probabilities of all “known” cosmological theories is. It could be 1% even, and it still won’t matter. The remaining 99% is irrelevant for arguing against creationism. Because only the relative probability matters.
Now, you can choose instead to argue not from the fact that known theories are (say) ten times more likely than creation theories, but from the fact that “unknown theories” sum to a prior a thousand times more than creationism. But then you have to do two things: (1) prove that “unknown theories” sum to a prior a thousand times more than creationism (very hard to do logically; it gets us back to my question above, as to how you know such relative probabilities, about theories you by definition know nothing about) and (2) prove that creationism doesn’t “better explain” the evidence by a thousand times in turn (hence, scenario (2) again, per my first reply, although at least here you can succeed by simply showing that P(E|unknown causes) = P(E|creationism), yet that is still, again, rather hard to do, for theories you by definition know nothing about).
As to your final point about the theistic argument ex nihilo, I quite agree on that. But that’s a very isolated case. I’ve already surveyed what’s wrong with it here.
I found the article I was referring to before. It was Paul Draper who defined the fallacy, and Lowder who wrote a blog about it:
Paul Draper, the Fallacy of Understated Evidence, Theism, and Naturalism
This illustrates what I was saying about the X –> E fallacy above. Theists do this all the time. One would do well to always catch it when they do.
“Kalam can be refuted in 60 seconds by simply pointing out there is no logical connection between it’s penultimate and ultimate conclusion (there is no syllogistic link between “there was a first cause” and “the first cause was a personal being,” that step is always a big hand waving trick) and therefore it’s useless. It can at best prove a quantum wave function started the time-space continuum, if we combine its penultimate premise with the best established facts of science.”
True enough, but I think most theists would argue that IF the universe had an absolute beginning, that means it must have arisen from “nothing”, and they would dismiss the quantum wave function hypothesis because that is already “something”, which in turn would require an explanation and is incompatible with the “absolute beginning” conclusion previously established. I’m never sure how to respond to this kind of argument, so usually I prefer pointing out that it has not yet been established that the universe had to have an absolute beginning, and thus have necessarily a cause. Perhaps you can elaborate (then again, I believe the problem lies on how you define “universe”).
On a side note, as far as I know, Wes Morrison’s criticism of the kalam is considered to be one of the most forceful refutation of this argument.
http://stripe.colorado.edu/~morristo/kalam-not.html
http://www.colorado.edu/philosophy/wes/metaphysical-time.pdf
http://stripe.colorado.edu/~morristo/craig-on-the-actual-infinite.pdf
http://stripe.colorado.edu/~morristo/infpast.html
http://www.colorado.edu/philosophy/wes/wes2craig1.pdf
http://www.philoonline.org/library/morriston_5_1.htm
http://www.colorado.edu/philosophy/wes/wes2craig2.pdf
(links found thanks to ex-apologist’s blog)
God is not nothing either. So the competing hypotheses are “God” or “quantum wave function.” The latter is vastly simpler and has vastly more background evidence in its support (indeed, it requires positing no new entities not already observed).
Any argument that “but God must have existed then” will also go to prove “but the quantum wave function must have existed then” or whatever else one proposes.
That’s why the Kalam lacks any logical syllogism that gets from “there was a first cause” to “that cause was a personal being.” Hence, it’s not a logically sound argument. It’s just hand waving.
But one can even keep beating the dead horse, since time cannot even in principle have a cause. Causation being by definition temporally prior to its effect–and if you abandon that and allow causes to be simultaneous to their effects, the first-cause argument goes out the window, since if simultaneity can obtain, then that entails just positing a first thing, which exists in the first moment of time and not before it, which eliminates any syllogism that gets to god as having to be that first thing. And then there’s the fact that we have no evidence that there was a first moment of time (the Hawking-Penrose theorem was refuted by Hawking and Penrose themselves decades ago). And so on.
Kalam is just nonsense on stilts. Anyway, my refutation summarizes all this and more in briefer space in the Carrier-Wanchick debate (search “KCA” here and here).
Audio of the debate is up at this point. See here.
Video, too.
Thanks. Yes! For the video now see here.
Interesting debate.
One thing that immediately caught my eye in Marshall’s comments was the extent to which some of his arguments depended on the reliability of personal testimonials. But unfortunately, we all know from the multiple examples provided by both young-Earth creationists and Intelligent Design creationists that Christian apologists frequently can’t be trusted to tell the truth. And anyone who has read Marshall’s book on the new atheism knows that Marshall in particular can’t be trusted to tell the truth.
Here’s just one example of his “truths behind the new atheism.”
1. Marshall falsely accuses Dawkins of being inconsistent in claiming that the search for irreducible complexity (IC) is both scientific and unscientific. (p. 63)
Marshall concocted this false accusation by taking Dawkins’ statements out of context. Dawkins’ actual argument is that IC is: (i) scientifically relevant in trying to *falsify evolution;* but (ii) scientifically irrelevant in trying to *prove ID.* Since Dawkins’ two statements relate to two different functions (disproof vs. proof) and two different theories (evo vs. ID), it’s obvious that they are no more inconsistent than reporting that the Yankees won their first game and lost their second game. Those baseball scores are not logically inconsistent, because they refer to two different contexts. Dawkins’ statements are not logically inconsistent either, for exactly the same reason.
Interestingly, when Marshall was challenged on this point in another forum, he defended his false accusation by concocting yet another falsehood, claiming that Dawkins had basically said that it’s OK for evos to point to biological structures that disconfirm evolution, but not for creationists to do the same thing. Marshall never provided a citation for this remarkable claim, perhaps because he knows it’s an outright falsehood, just as his original accusation was.
Incidents that make Marshall look like an unrepentant, serial liar.
A list of 30+ additional, blatant falsehoods or other misleading arguments from Marshall’s book on the new atheism can be found here.
http://www.amazon.com/forum/religion/ref=cm_cd_pg_pg1?_encoding=UTF8&cdForum=Fx1M9TK6UGAX6EO&cdPage=1&cdThread=Tx3UCB48O3G6426
Marshall has known about most, if not all, of these allegations for years, and yet as far as I know, he has not responded to a single one in any substantive manner. (He did refuse to provide a supporting reference for a couple of challenged statements, but I don’t think refusing to provide supporting documentation counts as a “substantive” response.)
In any event, when someone routinely resorts to blatant falsehoods and then turns around and argues for the reliability of personal testimonials, that doesn’t strike me as a very convincing argument.
mr carrier, i left a comment on your blog but it didn’t go through because i think it wasn’t relevant to the post
i left it here
http://www.richardcarrier.info/archives/3195
and it is still awaiting moderation
here is what i wanted to know
you recently had a debate with david marshall
http://debunkingchristianity.blogspot.co.uk/2013/04/richard-carrier-vs-david-marshall-is.html
you talked about hand washing and jesus
i will quote the relevant verses
1Then come unto Jesus do they from Jerusalem — scribes and Pharisees — saying, 2‘Wherefore do thy disciples transgress the tradition of the elders? for they do not wash their hands when they may eat bread.’ 3
And gathered together unto him are the Pharisees, and certain of the scribes, having come from Jerusalem, 2and having seen certain of his disciples with defiled hands — that is, unwashed — eating bread, they found fault
for the Pharisees, and all the Jews, if they do not wash the hands to the wrist, do not eat, holding the tradition of the elders, 4and, [coming] from the market-place, if they do not baptize themselves…
5Then question him do the Pharisees and the scribes, ‘Wherefore do thy disciples not walk according to the tradition of the elders, but with unwashed hands do eat the bread?’
15And Peter answering said to him…that all that is going into the mouth doth pass into the belly, and into the drain is cast forth? 18but the things coming forth from the mouth from the heart do come forth, and these defile the man
And Peter answering said to him, ‘Explain to us this simile.’ 16And Jesus said, ‘Are ye also yet without understanding? 17do ye not understand that all that is going into the mouth doth pass into the belly, and into the drain is cast forth? 18but the things coming forth from the mouth from the heart do come forth, and these defile the man; 19for out of the heart come forth evil thoughts, murders, adulteries, whoredoms, thefts, false witnessings, evil speakings: 20these are the things defiling the man; but to eat with unwashen hands doth not defile the man.’
me:
notice that only a religious issue is at stake and jesus SEES NO OTHER REASON for washing hands before meals?
notice that jesus could only SEE ritual within the practice and nothing else?
notice that he agrees that it is NOT IMPORTANT to wash ones hands? notice that the deciples with unclean hands would continue to eat with unclean hands because of jesus’ response to the pharisees? jesus gave them an escuse to keep thier hands dirty and continue to eat because according to jesus’ logic it is not IMPORTANT to clean hands before meals.
“Jesus never gave any good advice reflecting what is known now but was not known then. So Jesus should have been saying that it’s important to have clean hands to avoid contaminated or being contaminated, but not just to follow a ritual mindlessly.”
but then there a christians who say that it is not talking about “literally dirty hands” but…
i quote:
“the word translated as defiled or unwashed (you know, you put it in caps–that means you are right!) is defiled or unwashed in the ceremonial or purification sense. Likewise people were, for various reasons, told to ceremonially wash themselves not because they were literally dirty but because it was a purification right”
i don’t know anything about the greek and i have questions about this
can the greek word used refer to LITERALLY dirty hands?
Mark 7:1 The Pharisees and some of the teachers of the law who had come from Jerusalem gathered around Jesus and 2saw some of his disciples eating food with hands that were “unclean,” that is, UNWASHED.
so the pharisees OBSERVED literally dirty hands, right?
I’m not sure what you are saying. The facts are:
Pharisees assiduously washed their hands before meals.
Jesus’ disciples did not.
The Pharisees asked Jesus why.
Jesus said there was no reason to wash hands before meals.
Factually, Jesus is wrong.
Had Jesus known about germs he would have said everyone, his disciples included, really should wash their hands, and not because it was tradition, but because it would save lives and improve health.
Therefore, Jesus did not know about germs.
God would know about germs.
Therefore, Jesus was neither God nor in communication with God.
greetings mr carrier
i will first quote this translation:
“““Some Pharisees and teachers of the Law who had come from Jerusalem gathered round Jesus. They noticed that some of his disciples were eating their food with hands which were ritually unclean – that is, they had not washed them in the way the Pharisees said people should” (Mark 7:1-2)”
some christian aplogists assume that the hands were CLEAN , but NOT CLEAN in the way the pharisees wanted them i.e “ritually clean”
note in the above translation “… NOT WASHED THEM in the WAY THE pharisees….”
youngs literal
HERE IS YOUNGS literal
1Then come unto Jesus do they from Jerusalem — scribes and Pharisees — saying, 2‘Wherefore do thy disciples transgress the tradition of the elders? for they do not wash their hands when they may eat bread.’ 3
And gathered together unto him are the Pharisees, and certain of the scribes, having come from Jerusalem,
2and having seen certain of his disciples with defiled hands — that is, unwashed — eating bread, they found fault
for the Pharisees, and all the Jews, if they do not wash the hands to the wrist, do not eat, holding the tradition of the elders, 4and, [coming] from the market-place, if they do not baptize themselves…
5Then question him do the Pharisees and the scribes, ‘Wherefore do thy disciples not walk according to the tradition of the elders, but with unwashed hands do eat the bread?’
so does youngs translation support the claim that the HANDS WERE LITERALLY dirty? please explain
thank you
Ah, I see. You have been misled by specious bible translations. The relevant verse is Mark 7:2 which reads in Greek:
Which translate literally as:
The key word “unwashed” is aniptos (likewise again in Mark 7:5). It means what it says: unwashed.
The phrase “in the way” is nowhere in the text. Even the next verse reads only:
You will see no reference to the words your translation inserts. Those words simply aren’t there. The phrase “to the wrist” is a modern attempt to interpret “by fist” [dative of pugmê], the more direct meaning of which is that they wash their hands with their fists (i.e. the way we scrub our hands, enclosing one in the fist of the other), meaning they wash well. Note that the disciples are not said to have washed less well, but to not have washed at all.
Hence what is being described is simply washing their hands, which “some of the disciples” weren’t doing–their hands were “unwashed” (notably, the tradition Jesus goes on to denounce here included washing your cooking and drinking utensils, too, cf. Mark 7:4, another obvious vector for germs that Jesus was evidently unaware of).
thank you mr carrier.
Hi,
You said at the debate that Joe Nickell had some good books to read. I looked those up and it seems that Nickell has made almost 30 books. Would you like to recommned some of them that you have read? Of course, I’m interested about the books that talk about religion or miracles.
Select the book(s) that discuss the general topic(s) you are most interested in (e.g. Entities for ghosts etc.). But in general there is no way to force rank them. They simply represent a collection of all his investigations over several decades. It’s basically like one multi-volume book.