« Engelbart: Insufficiently Recursive | Main | Total Nano Domination »

November 26, 2008


The pope believes his religion is God's word. So the whole idea of having a discussion with people, and at the end of the argument their opinion overriding God's, must seem ridiculous to him.

>...the sorts of things that actually justify beliefs (evidence, argument)

I think this is where the divergence of opinion is. The Pope sees faith as a different source of beliefs. Assuming the Pope is an honest catholic, he considers his (and his follower's) faith superior to that of all other religions. To him, "everyone else just doesn't get it".

I disagree with you concerning (1), for the following simple reason: you and the pope have different utility functions.

Your utility function looks like this:

YourUtility = Present Utility + discount * Future utility

To maximize this, you need the best possible estimate on Future Utility. This depends on both future events AND your future self.

The utility function of an ideal Pope looks like this:

PopeUtility = PresentServiceToCatholicGod + discount * FutureServiceToCatholicGod

By engaging in debate, the pope may alter his utility function. This does not optimize ServiceToGod in the present or future! The Pope's future self is irrelevant to his utility, only logistical information is useful (e.g., where missionaries will be most useful).

Or, (5), pope is saying stuff that's more likely to preserve the blind faith of his followers, regardless of the pope's actual beliefs (or lack thereof).

Experience indicates that people in religious power tend to be smart, cynical, and good at maintaining an image.

The Catholic church sees the church as a body rather more literally than protestants do. They divide up responsibilities among different roles. In particular, thinking critically about religion is a responsibility given only to a small number of priests (basically, the Jesuits). Critical thinking, in Catholic tradition, is something dangerous and a bit embarrassing, and not done in the presence of the Catholic laity, let alone infidels.

Several readers of OB have expressed support for this approach (at least 4 as I count), when we substitute "Singularity Institute" for "Catholic Church" and "CEV" for "Christian doctrine". The reasons they give are similar to the reasons that the Catholic hierarchy has for not subjecting their faith to critical analysis in public. So perhaps the Pope is being rational.

Catholic philosophers have come up with a large literature rejecting basic 2. They claim to have gerrymandered non-Occam priors backed by gnosis, facing off with Robin's origin dispute challenge through a claim that if their prior is right God has selectively provided the gnosis. It's very similar to zombie-ists who think it astronomically unlikely that any given brain like theirs (on Occam grounds) would be nonphysically conscious, and yet think that they are nonphysically conscious.

These are great questions to consider. What is the nature of religious faith? How should a religious person view his reasons for his belief?

I would suggest a 5th alternative, a variant of your numbers 2 and 4: he believes his beliefs are held for reasons, and those of other people are also held for reasons, but he believes his own reasons are much stronger than those of other people.

That is why he says he doesn't want to put his faith (i.e. his basic beliefs) "in parentheses". He views those beliefs as fundamentally extremely sound and is convinced that abandoning them will lead to error. He sees interfaith dialog as requiring speaking hypothetically as if his beliefs are wrong, i.e. putting them in parentheses. But since he is so convinced of the soundness of his beliefs, even such hypothetical speech is at best pointless since it will automatically lead to wrong conclusions. And perhaps he is worried that for the pope to speak even hypothetically about catholicism being wrong will send a message that he sees a real possibility that it might be wrong, which would weaken other people's belief in catholicism and therefore lead to much more error and bad consequences in the world.

I actually find this to be a very common attitude (in at least a milder form). It is IMO at the root of almost all disagreements. When you disagreed with Robin's position on the nature of disagreement, wouldn't you say that this described your position? You knew Robin had reasons for his belief, but you had reasons for your own belief, and you believed (or at least assumed) that your reasons were better? Otherwise you would have changed to his belief.

Any medieval philosophers here? It seems that when it comes to Catholicism and Reason, the Middle Ages is where the action is. And there are a lot of wrinkles in there that I'm pretty sure the Catholic church still buys.

Aquinas said that there are two types of knowledge - that which can be revealed by reason and that which can only be revealed by faith (which can be interpreted as 'directly from God' if one is so inclined). That which is taken on faith is an axiom - trying to verify axioms using reason is wrongheaded. If you are in a dispute with someone who is holding different axioms, then reason will not help - the only way to resolve that is to agree to use the same axioms (and work within the same formal system). Since that would be hypothetically rejecting the word of God, the Pope understandably doesn't think it's a good idea.

Carl and Thom are right. An irony of history is that one of the key figures in the Enlightenment was Pierre Bayle, who argued at length that human reason was demonstrably unreliable because it would cause us to reject the Bible.

In defense of the Catholic tradition, I think (could be wrong) that by "reason" they meant Aristotelian logic. Today, we know that a lot of reasoning doesn't fit the Aristotelian mold. Back then, anything that didn't was probably lumped under "experience", which was and is regarded as a different valid basis for belief.

A major disagreement between the Catholic Church and us "rationalists" is that the Catholic Church takes "you can't agree to disagree" seriously. Many Catholics, especially in the middle ages, believed that, since the dead outnumbered the living, the opinions of the dead should outweigh the opinions of the living.

Now that the living outnumber the dead (I think), the Catholic Church should reduce the importance attached to tradition.

I see that Hal Finney has expressed what I had vaguely formulated, only much more clearly.

Also, we might want to investigate what the Pope meant when he said "in theological terms" preceeding "true dialogue". I suspect he meant something at least slightly different from what Mr. Gowder uses.

(I suspect he might refer to the sense shown here under "The dialectic method in theology". His wording has, to me, the feeling of being Terms of Art that I think Mr. Gowder might not have fully appreciated.

This is a common difficulty in discussion of theology in the Catholic tradition by those without eg. a Jesuit education - myself among them.)

I don't know what the Pope would say, but many Christian apologists object to the premise mentioned in #2 "that his beliefs and the beliefs of, say, Muslims, are roughly on an epistemic par [...]." Their argument is that Jesus's followers eventually became convinced that he had materially risen from the grave (a fact that needs explaining somehow), while the Muhammad tradition claimed that he performed no miracles apart from the Qur'an itself. In isolation, this might induce a shift in posteriors toward Christianity of at least a factor of 2 or 3. The accuracy of the other claims of each religion would need to be evaluated, but it's not obvious to me that it would tip the scales one way or the other, especially since the Qur'an appears to endorse most of the problematic stories in the Bible, including the Garden of Eden, etc.

Phil Goetz:

A major disagreement between the Catholic Church and us "rationalists" is that the Catholic Church takes "you can't agree to disagree" seriously.

Well, speak for yourself! Some of us "rationalists" take the disagreement theorems seriously!

Many Catholics, especially in the middle ages, believed that, since the dead outnumbered the living, the opinions of the dead should outweigh the opinions of the living.

I had not heard this, that is quite interesting. But it seems odd since in no era have the majority of the dead been Christians.

And maybe it's not as crazy an idea as it sounds. I wouldn't take the word of the dead on everything, of course. But I would not be surprised if, among the relatively few issues where the dead would disagree with the living, the dead turn out to be right on more of them than many people would expect.

Now that the living outnumber the dead (I think), the Catholic Church should reduce the importance attached to tradition.

Actually it appears that the dead still have the edge in numbers, outnumbering the living by more than 15 to 1. However this is in large part because infant mortality was very bad, hence birth rates had to be much higher than today. So the majority of the dead would be children, who might arguably not be fully rational actors whose opinions we would count equally with adults. However I doubt that children outnumbered adults among the dead by a large enough ratio to outweigh the numerical advantage of the dead over the living.

Basic 2 as stated is untenable, unless you take belief support structures to be ultimately circular.

  1. Ostensibly you believe that a reason is a belief that supports another belief. If I believe q on the basis of p, then p is a reason for q.
  2. Ostensibly you believe that if p is a reason for q, then it isn't rational for q to be my reason for p. That is, rational support relationships between beliefs can't be directly reflexive. (This is not to say that, in a rational belief support structure, q can't ultimately be taken as strengthening p; it's just that if p is a reason for q, then there must be a sufficient set of reasons for p that doesn't include q.)
  3. Ostensibly you believe that it is implausible for a human to entertain an infinite number of beliefs; hence, any belief support chain must stop somewhere, or be circular.
  4. Hence, any belief a person P holds in a rational belief support structure must ultimately either (a) be part of the finite set of beliefs held by P, or (b) be unsupported by other beliefs that P holds; it must be what a foundationalist might call a "basic belief".

Let's take your Basic 2. Doubtless this qualifies as one of your beliefs. You hold the proposition expressed by Basic 2 to be true. But if you believe it's true, you must have a reason for that belief. And you must have a reason for the reason for believing Basic 2. And so on. Unless the chain of supporting beliefs in your belief support structure is infinitely long (and I contend that this is terribly implausible), the chain will either terminate in an unsupported belief, or it will wind its way through your finite rational belief support structure without end, in which case your belief support structure is ultimately circular.

I think you either must deny one of (1)-(3), or you must affirm one of (4a) or (4b). Personally, I affirm (4b): I believe that there are certain beliefs that it is proper to hold without rational justification (not that we think these beliefs aren't grounded in ways that can be understood rationally). For instance, the person who has never reflected on the reliability of her sense perception, and who can thus have no non-circular rational justification (conscious, sufficient reasons) for trusting her sense perceptions, is nonetheless in most circumstances within her rights to affirm the beliefs she forms spontaneously and without conscious thought based on them. The same holds for large classes of memory beliefs and beliefs formed on the basis of the testimony of others. Consider that the vast majority of beliefs about scientific matters that even scientists entertain are held based on the testimony of their colleagues and not based on their own work. No one could function, within science or without, without accepting a great many beliefs on the basis of the testimony of others. Nor is it irrational to do so.

One further point: neither (4a) or (4b) is much help in dealing with the Pope, who is by all accounts an exceptionally able theologian. I take it that it's likely that his belief support structure is as coherent as yours and perhaps more so. Under (4a), coherence would be one of the chiefest epistemic virtues, and his belief support structure likely enjoys that virtue to a high degree. How would you argue against his position?

Perhaps you affirm (4b): there are some beliefs that are properly regarded as basic: we may hold these without support from other beliefs (even if we may find rational grounds for them ex post facto). In this case, the justification (or perhaps better: warrant) enjoyed by beliefs such as the Pope's will turn not on epistemological matters but rather on the question of whether or not those beliefs are true. In the Christian tradition that looks back to Augustine, Aquinas, and Calvin (and, I would say, to the Bible), it's pretty clear that the account of human nature as created by God entails that God in some fashion has either provided us with a latent knowledge of his existence or provided us with a faculty that leads us to form (true) beliefs in his existence under certain circumstances. That is to say, if Christian belief is true, then it is almost certainly warranted; if it isn't true, then Christian belief is not. If the God I worship exists, then my (true) belief in him does not come down to epistemic luck. The epistemological question depends on the ontological question and is not independent of it. In the case of (4b) and in the case where Christianity is true, beliefs in the Christian God formed under at least some circumstances are properly basic.

If you want to argue against Benedict on epistemological grounds, you must take a stand on some basic epistemological issues. Really, if you want to make an informed argument, you ought to interact with recent analytic epistemology of a Christian bent: Alston, Plantinga, Wolterstorff, and others.

Melchior has presented Plantinga's fully general defense. A religion may look precisely like a human fraud, and be repeatedly shown to be such, but if the religion makes further fraudulent claims of a sensus divinatus, one is "within one's rights" to believe it.

Nonsense Utilitarian. Many call themselves Christians and deny the resurrection. Many call themselves Muslims and assert that Mohammed ascended bodily into heaven or did other comparible miracles. Also, the Qur'an endorses most of the ethically problematic stories in the Bible but rejects most of the logical difficulties such as a god who needs to rest, calling attention to them as it does so.

"I take it that it's likely that his belief support structure is as coherent as yours and perhaps more so"

I don't see the support for this. The Pope has clearly faced dilemmas such as this

Age of the Earth as "revealed by reason" = 4.5 billion years
Age of the Earth as "revealed by faith" = ~6,000 years

and has either rejected reason or accepts that the biblical account is not correct. In either case he has abandoned any semblance of a "belief support structure." He believes simply because he wants to believe.

melchior, I assume "foundational belief" is just another name for "prior"? If so, I interpreted Paul as defending Robin's argument for common priors. Paul said, "This includes beliefs that are based on un-swamped priors from, e.g., how one was raised."

Greek philosophy's view on 'faith'

"In the teachings of the Bible, in contrast, it is necessary to have faith, utter and uncritical confidence, a conception quite alien to the spirit of Greek philosophy.*"

"* For classical greek philosophy, as for Plato, faith (pistis) is the lowest form of belief, characteristic only of the wholly uneducated, who fail to reflect critically on what they experience or are told. The Jewish-inspired Christian emphasis on faith struck educated pagan observers with astonishment; it represented, in their eyes, the extreme of anti-intellectualism - 'foolishness'."

From page 78 "The Perfectibility of Man" by John Passmore (1970)who is Professor of Philosophy at Australian National University. It is well worth a read if you are interested in the history if ideas (including Catholicism and Reason)

What's the point in talking about Catholicism if you don't know anything about it?

One might reinterpret religion to strip away the propositional content, but then one loses everything that makes religion different from any random social activity.
This couldn't be wronger. The propositional content of religion is the least important thing about it. If propositional content was all there was to religion, science would have vanquished it long ago.

You know, there is a vast body of knowledge on the sociology of religion. Since this blog is dedicated to big T Truth, you might consult some of it rather than making up stuff just because it sounds good.

I think there are tricky issues in cognitive architecture that make one expect beliefs without reasons. Consider the following naive architecture for an artificial intelligence: it has a database and an inference engine. A well know problem is that a contradiction implies anything. Any contradiction becomes a time bomb. Deductions burn fuse. Eventually the contradictions start showing up everywhere and the AI suffers a total credulity crash.

The problem is a problem with logic itself, which is brittle. Any form of intelligence needs hardening against contradictions. So all functioning intelligences probably "implication barriers" and "chain length limits" to stop contradictions propagating and crashing the system.

In mammals, thinking seems bolted on, on top of a short term reward system, and beliefs have emotional impact; that is part of how the brain works. What happens to beliefs that are within implication barriers?

The thinker cannot adjust them to correspond to reality because he does not draw their implications. That means he cannot check their implications against reality. On the other hand he doesn't act on their implications, so there is a certain freedom of choice, due to lack of consequences for worldy action. One expects the thinker to chose the beliefs that are most emotionally satisfying. Religion is about wireheading the implication barriers.

One point in favour of this theory is that it solves the puzzle of the buggered altar boy. The puzzle is that the sexual misconduct is committed by some-one who thinks that they face eternal damnation for their sins. How does that work? If you really believed in eternal damnation and you were feeling temptations that were gradually gaining the upperhand, you would have a nervous breakdown.

The answer is that the belief is behind an implication barrier. It is pleasant, perhaps, to believe that ones meagre rewards in this life will be topped up in the next and that your enemies will answer for their crimes, if not now, eventually. But one does not draw implications. One does not think: I know, I'll commit suicide and get to heaven sooner,... Nor does one think: if I give in to tempation, and since I know that people who give in to temptation are damned, then the implication is that I will be damned too.

Realizing beliefs require reasons is hard. As in, humanity stayed a few millions years before starting to suspect it. Most of us have a gut feeling wishful thinking works. Before I read Eliezer's sequence on evidence as interaction causing correlation and minds as engines and knowledge as negentropy, I wasn't fully conviced beliefs required evidence. And it's long and hard to explain, due to huge inferential distances. A thousand dollars say the pope doesn't really think beliefs require reasons - he may or may not say it, but he nudges his probability assesments without actual evidence. So why should he act like his beliefs needed reasons?

Manon de Gaillande: regarding that thousand dollars, what, in theory, would you consider to be evidence that the Pope does think that belief requires reasons?

And in any case, no human being has well calibrated probability assessments, and so in this sense there is no human being who does not "nudge" his probability assessments without evidence, whether he wishes to or not.

One point in favour of this theory is that it solves the puzzle of the buggered altar boy. The puzzle is that the sexual misconduct is committed by some-one who thinks that they face eternal damnation for their sins. How does that work? If you really believed in eternal damnation and you were feeling temptations that were gradually gaining the upperhand, you would have a nervous breakdown.

Isn't that what "forgiveness" is for? Merely confess your sins and repent, and all will be forgiven...

Forgiveness is a nice thing to believe in provided you don't follow through the implications. You are enjoying believing that your enemies will meet a sticky end due to not being forgive. You are enjoying believing that God will let you off with your own sins. This had better be happening in a mental ghetto in which dots get left unjoined or its not going to deliver emotional satisfaction.

Alan: forgiveness is supposed to happen only if you repent. Since when do your enemies repent?

The comments to this entry are closed.

Less Wrong (sister site)

May 2009

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30