site banner

Rationality and Religion

1 Introduction

In the Small-scale Questions thread, @TheDag asked:

[H]ow do you handle the paradox of belief? [...] The 'logical' part of my brain relentlessly attacks what it sees as the foolishness of religion, ritual and sacrament. And yet, when I partake and do my best to take it seriously, I feel healed. [...] How do you make sense of a serious religious practice, while keeping the ability to be seriously rational?

This post is my attempt to answer that question.

My apologies in advance for any first-draft typos or errors.

I am an Orthodox Christian -- a convert to Orthodoxy, but not to Christianity in general. I've been reading material from LessWrong/SSC/ACX for about 10 years now, but never considered myself a Rationalist, in large part because of the movement's basically-axiomatic rejection of anything not comporting with a materialist metaphysics. Nevertheless, I'm a natural skeptic and a mathematician by training, and I think I understand, at a visceral level, what TheDag is talking about.

This post is not intended to be an apologia for Religion, Theism, or Orthodox Christianity in particular. Instead, it is an outline of my way of thinking about Reason and Christianity, and why I think that (some forms of) religion -- yes, serious, supernaturalist, actually-believe-the-creeds Christianity complete with ritual and sacraments (in fact, especially that kind) -- is fully compatible with being rational; at least, as rational as we can reasonably expect to be.

Small disclaimer: I'm going to use Christianity, and (sometimes) Orthodox Christianity in particular, as my source of examples/topic of discussion. I (a) do not guarantee that everything I say will be precisely correct Orthodox doctrine (I'm doing my best but I'm not getting feedback from a committee of bishops and theologians) and (b) don't know how applicable this all is outside of Christianity. (It would be kind of weird if I thought that Christianity and other religions were in exactly the same position, since I think Orthodox Christianity is true and other religions varying degrees of less-than-true.)

2 The Goals of Rationality

Why does anyone care about being rational in the first place? The usual answer, which in my opinion is basically correct, is that there are two reasons:

  1. Because it helps you to believe true things rather than false things. ("Epistemic Rationality")
  2. Because it helps you make better choices. ("Instrumental Rationality")

Note that these goals are just that -- goals. There's no law of the universe (at least, there's no non-circular argument) that a particular "Rational" way of thinking will always be the best way to achieve those goals. A particular set of scientific, logical, and probabilistic methods seem to be pretty good, overall, and certainly excel in some domains, but in principal these are secondary to the above goals. Do you want to believe true things and live well, or do you want to Be Rational? Obviously the first, right?

Well...

There's another kind of reason to want to be rational. Maybe you have a skeptical temperament, and have an internal demand for a certain sort of rigor. Or maybe you have developed a kind of self-identification as a Rational Person, which has attached itself to a certain set of assumptions and ways of thinking. Or maybe you like to think of yourself as Intelligent and Rational, and there's this bunch of intelligent people you know, and they all say that thinking in a certain way, and believing in a certain set of axioms, is a prerequisite to being Intelligent and Rational, and theism and rituals and faith and religion is just Dumb Stuff for Irrational People and you don't want to be Dumb and Irrational, right?

(It should go without saying that this is a general You, not about TheDag in particular, but here I am saying it anyway.)

The important thing here is that these temperamental, identity-based, and social reasons for wanting to Be Rational are not, themselves, rational or virtuous. If it's the identity or social reasons that have got you, all I can say is that the faster you admit it to yourself and work on getting rid of them, the better.

But perhaps your troubles are in part due to a skeptical temperament, whether natural or trained, or with a difficulty believing that doing and thinking in ways that are not Rational could possibly lead to believing true things or living well.

In that case, the rest of this essay is for you.

3 Ontology

Some people are Christians because they trust authority figures who tell them it's true. Others are Christian because they believe they've witnessed an inexplicable miracle. There's nothing wrong with these people; many of them are better people than I am; but they are not me.

I am a Christian because of the Hard Problem of Consciousness.

Okay, maybe that's a bit too glib, so let me expand a bit. There is a fundamental mystery of how consciousness can exist in a purely material universe. I don't mean that it's a mystery how something could exhibit intelligent behavior, or have some sort of internal model of the world that contains itself. I mean that the existence of a first-person perspective, of there being an I that sees from my eyes and thinks my thoughts, of there being a quality to experience -- all things that we take for granted -- seem impossible in a materialist ontology. The usual materialist takes either handwave the problem away, or else (inexplicably to me) bite the bullet and deny the existence of the conscious self at all.

Even so, I exist.

Lest I digress into the apologia which I did not intend to write, let me just make my main point here: the existence of a first-person perspective not only reveals materialism to be a premise rather than a conclusion, it poses a problem for the universal applicability of rationality, because while the first person perspective is a universal and undeniable fact, even the best thinkers cannot seem to articulate what, exactly, it is, or delineate it to the point of being able to reason clearly about it -- which is why we see the problem being dismissed as just muddled thinking by others.

My other point in bringing this up is as a segue into talking about exactly how deeply the Theist (or at least, Christian) ontology differs from the Materialist one. A lot of people have this unspoken idea that Christian ontology is essentially the same as materialist ontology, except that there is are extra entities which maybe don't follow the laws of physics, and one of them is "omnipotent" (whatever that means, maybe power level = infinity or something), and we call that one "God".

This is not the Christian ontology.

The actual Christian ontology is something more like this: The fundamental nature of reality does not look like atoms and the void, governed by laws of physics. Rather, the fundamental nature of reality is something which is in most respects unimaginable, but in which what we call personhood and will and morality and love and reason are fundamental attributes. This is God -- not another entity like a star or a chair or a cat or a human, only immaterial and superpowered, but rather, the Person at the heart of all reality, in virtue of which everything that exists (including, of course, the entire material universe and all its physical laws), exists.

This is so fundamentally difficult to get one's mind around that people resort to paradoxes to talk about it: We call God "The Existing One", and yet some Christian theologians have said things like "God is not a being" -- not because they think that God is just some idea, but because our notion of "existence" or "being" imports the idea of a separate entity within the universe, and is insufficient to what -- who -- God is. (More on this in the next section.)

This ontology is probably shocking to people whose habitual assumptions are materialist -- which is true of most people, let alone Rationalists. So they round off theistic claims, in their head, to something like "Superpowered Invisible Man". This concept is, from the Christian perspective, nearer to the truth than pure materialism, but -- the skeptics are right on this one -- being materialist-except-for-this-one-superpowered-dude is not very rational.

But within the ontology I've outlined, Christian beliefs about the world make reasonable sense -- I would say they are rational, not in the sense of being obviously inevitable or circumscribed by reason, but in that they don't pose any problem for a rational person who recognizes his limits and is content with partial understanding.

4 Cataphasis and Apophasis

When people talk about paradoxes in Christianity, they generally mean one of four things:

  1. Doctrines, like the Trinity, which refer to concepts that our minds have a difficult time comprehending, because they are so different from our usual experience and categories.
  2. Counterintuitive truths, expressed in apparently-contradictory language in order to draw attention.
  3. Deliberate paradox in the form of Apophatic theology, meant to explode misconceptions about God and emphasize our inability to comprehend His fundamental nature.
  4. Multiple ways of talking about the same topic that seem to be inconsistent.

Of the second I will have nothing further to say; it is clearly not a problem for rational thinking. Of the first, I want to emphasize that the apparent paradox is due to our inability to understand the concepts involved and nothing more, much like how arithmetic on infinite cardinal numbers is not a "real" paradox just because it doesn't behave like arithmetic on the integers. ("But I understand cardinal arithmetic, down to how it is a consequence of ZFC! If nobody understands the Trinity fully, how could it be reasonable to believe it?" More on that later.)

So let's talk about the third and fourth.

A number of foundational Christian thinkers have divided theology into two parts: Cataphatic, or positive, theology, and Apophatic or negative, theology. Cataphatic theology is what is at play when one says things like "God loves", or "God is merciful", or "God is just"; or that which is expressed in creeds and dogmas. Cataphatic theology is saying the things that we know about God. Apophatic theology is an approach in which, rather than making positive statements about God, we make negative statements about what God is not. (For some easy examples: "God is not material", "God does not have a cause outside Himself".)

Apophasis often takes the form of paradox when juxtaposed with cataphatic statements, because, first, our concepts which are employed in cataphatic statements will smuggle in implications or impressions which are not true, and second, because this paradox emphasizes our inability to comprehend the full truth about God. I mentioned the apophatic "God is not a being" above, for instance, which seems to contradict theism, but actually the point is that our notion of "being" or "existence" is not really applicable to God.

One might think of apophatic theology's relationship to cataphatic theology as trying to help us understand the "map" of cataphatic doctrine as a guide to the "territory" of who God is and how we relate to God, by continually pulling our attention to the fact that the map is not the territory. This isn't irrational paradox at all, but our continual reminder that the person at the center of reality is not something we can really get our minds around, and we're better off not imagining that we can.

(Digression: Apophatic theology is not unique to Christianity; there is something very similar in Neoplatonism as well as, I think, in Taoism ("The Tao which can be spoken is not the true Tao.").)

Finally, the fourth kind of paradox. It is much like the third, except that multiple counterbalancing positive statements are made, each pointing to part of a truth which is too difficult for us to really get our heads around. Now of course it is possible to excuse nonsense as "just different aspects of an incomprehensible truth," but the thing can really happen as well as being faked.

Let's take an example: What's the deal with sin? Why is it bad for me to sin? (other than it being bad for the people I harm)? The following answers are all defensible from both the Bible and Christian Tradition:

  1. Sin is breaking God's rules. It makes God angry, and He will punish you for it. (BUT: Doesn't the Bible also say that God hates no one and is quick to forgive?)
  2. Sin is bad because it's foolish, and tends to lead to bad natural consequences: material, psychological, or social. (BUT: People who do bad things often end up ahead.)
  3. Sin is like a progressive illness; if you sin, you get sicker, and eventually you'll be miserable (unless you get cured). (BUT: where's the will and personal guilt in all this? And why do I need to consent to being cured?)
  4. Sin separates you from God, and the absence of God's love ends up in misery. (BUT: How can anyone be separated from God and God's love, if God is everywhere and in everything, and loves everyone?)
  5. Sin breaks your relationship with God (BUT: a human's relationship with God is only similar by analogy to our relationship with other humans, and how could this be broken, since God doesn't get emotional baggage like humans do?)
  6. Sinning makes you into the sort of person that finds the presence of God intolerable. (BUT: how does that even work?)

(I probably left some out.) For what it's worth, I -- and many Orthodox theologians -- think the last one is probably closest to the truth, but in some ways it's the least actionable. What we get is all of them: partly because each of them is the right model for some occasions, and we, being unable to really understand the underlying reality, need a multiplicity of models for different circumstances. "All models are wrong, but some are useful," indeed.

5 Those Who Have Not Seen and Yet Have Believed

This section title refers, of course, to Jesus's words to the Apostle Thomas -- after the resurrection, Jesus appears to the Apostles, but for some reason, Thomas isn't with them. The rest tell Thomas, but he -- being a bit of a skeptic -- refuses to believe unless he can verify it for himself (down to unfakeable physical proof). Later, Jesus appears to all of them, offers that proof to Thomas -- and then gives a blessing to "those who have not seen and yet have believed".

There is an epistemic issue -- two, maybe -- that a lot of rational/skeptical people have with Christianity, and it's this. A lot of Christian doctrine contains claims that cannot be verified by anyone alive today (e.g the Crucifixion and Resurrection), or even could not have been directly verified by human observation at all (e.g. the Trinity).

The first is not, in principle, a problem. Everyone believes lots of things they can't verify, even things that nobody can verify now (historical events, e.g.), because they trust in the body of people who did observe those things and those who have passed on the report. They are not wrong to do so! Very little can be empirically verified by an individual. So part of the question, then, is how trustworthy are the people who reported and passed down these events? Since this is not an apologia I won't get into the weeds here (and also I'm not really an expert), so I'll just say that I think a good case can be made that the answer is "Pretty darned trustworthy, all things considered". Still, some of the claims made are pretty wild (cf Resurrection) if you haven't already accepted the overall metaphysics, so skepticism is understandable.

The second is more of a problem. How can anyone, no matter how honest or intelligent, come to know something like the doctrine of the Trinity, which is (a) something that can't be (physically) observed, and (b) admittedly not fully comprehensible by anyone? Christianity, of course, has an answer: it was revealed by God -- through the words of prophets, or Jesus, or by a revelation given to some of the Apostles. That's an explanation, but it has one problem: it does not bridge the epistemic gap for those who don't already broadly accept Christianity.

Here's the thing: this is fine. Nobody should be asked to accept these things just on the say-so of people they aren't sure they can trust. It is not rational to do so, but it's also not necessary. There are good ways to bridge that gap, such that blind belief is not required.

Roughly, it works like this: you get good evidence, of some sort, that at least some of the claims are true. Since all these claims are coming from the same source, they are tied together -- belief in one should increase your estimation that the source is a good one, and thus that the others, which you can't verify, are true as well. Coming to believe in the others to an extent, you see how they fit together (and/or find that believing other claims has good results). At some point a threshold is passed, and you believe not in the truth of this or that statement, but in the whole edifice, even those parts you don't understand (yet), because, as Chesterton put it, you find that Christianity is a truth-telling thing.

Talk to most thoughtful Christians, including many converts, and you'll find that something like this is the process. Maybe they have, like me, some deep philosophical convictions that turn out to be elucidated best by Christian doctrine. Maybe they had an experience that, while maybe not communicable to others, they feel they had no choice but to accept as miraculous, and which pointed them in that direction. Maybe they just found that acting as though the doctrines are true had good results for them that they did not find elsewhere.

As an exercise, I invite you to think about why, from the Orthodox Christian perspective, correct doctrine is so important. It's not because the beliefs, in themselves, are going to save someone ("Even the demons believe -- and tremble!"), nor the converse, that one cannot be saved without specific beliefs (see: the many saints who made errors or lacked knowledge, or the fact that the Church believes that children and idiots can be saved). It's not an arbitrary test, either. Rather, the Church believes that knowing certain truths about God and Humanity's relationship to God helps you, because God is real, and believing true things makes it is easier to be aligned to that reality, which is the real goal.

[ I ran out of characters, so the rest will be in a reply to this post.]

18
Jump in the discussion.

No email address required.

I appreciate the nuanced take and attempt to clarify your beliefs in a manner clearer to Rationalists and rat-adjacents.

Further, I wholeheartedly endorse the claim that rationality can be applied to any goal or belief system, that's what the orthogonality thesis is all about. There's nothing irrational about say, wanting to covert the universe into paperclips, and for all the insults I could hurl at a Paperclip Maximizer as it disassembled me for parts, irrational isn't one of them. It would only be so if it went about its goals in a counterproductive manner, or a grossly suboptimal one, such as, idk, building flawed Von Neumanns that make safety pins instead.

Thus, in principle, you can reconcile epistemic and instrumental rationality with religion, my point is that in practise it's about as sensible as Muslim apologists arguing that hijabs are feminist and empowering for women. Less gross hypocrisy or maybe even none on your end, of course, but the same issues apply.

If you end up in a situation where you are initialized with a prior of 1 in the existence of God/The Truth of Christianity, then you are perfectly rational in assuming all observed evidence in that light and then assuming any discrepancies or paradoxes arise from an incomplete understanding of the world.

It's still a deeply malign prior.

I don't know if the human brain, while strongly likely to be doing Bayesian Predictive Processing, is capable of having priors of literal one or zero. That being said, the religious seem close to the former.

All else being equal, a simple hypothesis or prior should be privileged over a more complex one when they are equally as good at explaining the evidence, or predicting the future. That is a basic consequence of probability theory, complexity needs to be justified. We only use the more abstruse equations for GR over simple F=ma and other classical physics because in certain very important regimes, they justify the headache and predict the world with more accuracy.

As it stands, the world as it exists today is far better explained by assuming the universe works solely according to the laws of physics without any external intervention where anyone can see it.

The Hard Problem of Consciousness is a sign that our existing knowledge and theories are insufficient for the task of explaining everything, but that isn't a blank cheque for believing in anything you feel like. The rational thing to do is remain agnostic on that regard, at least unless we end up with a Grand Unified Theory of Everything and yet no answers. The history of humanity is riddled with questions that seemed beyond the ability of empirical studies to explain, such as the elan vital, yet we have solved them, and Christian metaphysics is not a satisfactory explanation unless you start out with an unshakeable belief in it.

The existence of a Creator with the properties of Omniscience, Omnipotence and Omnibenevolence need justification far more powerful than the word of some Middle Eastern randos several millennia ago, and it's such a shame that more objective evidence has been sorely lacking since we have omnipresent digital recording devices and randomized control studies.

Was Jesus a real historical character? Easily could have been, but that no more makes a difference than the next schizophrenic I treat who rambles about being the next prophet.

Perhaps the strong relationship between visions and speaking in tongues with something as prosaic as temporal lobe epilepsy should make the wise take a hint.

Your arguments are sound if and only if one assumes the truth of your faith as given and then works backward to justify everything else. When you scramble constantly with apologetics against all other evidence, the intelligent thing to do is consider whether or not the null hypothesis has a point!

The mere existence of complex maths or seeming paradoxes is no excuse for waving away far bigger ones with less reason to assume that the reason that the properties you ascribe to them are anything but a consequence of them being imaginary.

As I was telling @Meriadoc yesterday, if I ever meet the Omnibenevolent loving Creator who created ichthyosis vulgaris, I'll kick them in the Holy Nuts. Until then, my sheer disdain for Him is made obviate by the fact that He's fictional.

Shame, because if prayer cured cancer or brought about post-scarcity, I'd be the busiest beaver in that regard. Beats going through med school and memorizing all these doses I tell ya.

Until that happens, I'm content to watch us make our own gods, not that they should be worshipped or programmed to demand it. Maybe Lucifer would be better served creating a Heaven on Earth instead of reigning in hell, at the very least we have the means of doing it ourselves.

if I ever meet the Omnibenevolent loving Creator who created ichthyosis vulgaris, I'll kick them in the Holy Nuts. Until then, my sheer disdain for Him

Do you hold it as a general belief that such a Creator could not create a physical universe where even a single bad thing happens? Which part of your materialism or rationalism does that belief come from?

It doesn't confirm materialism and rationalism are correct, only that the fact the Creator is omnipotent, omniscient and omnibenevolent. He can be two, but not all three. If you want to be a Spinozan and just call the materialistic universe God, be my guest. It's a hypothesis that predicts nothing and everything.

I did not ask whether it confirmed that materialism and rationalism are correct. I asked which part of your materialism and rationalism the conclusion that an Omnibenevolent loving Creator could not create a physical universe where even a single bad thing happens came from.

Pretty far goalpost movement from "no horrible viruses/genetic diseases" to "not a single bad thing".

Just don't create the real bad non-fundamental shit, bro.

Not a goalpost shift at all. Instead, I'm just trying to fill in the details of the argument and trying to figure out where his goalposts are. There could be other things in the middle. Like, perhaps you're proposing:

  1. Ichthyosis vulgaris exists.
  2. ??? [something coming from reality]
  3. Such a Creator could not create a physical universe where "real bad non-fundamental shit" exists, however that term in quotes is defined in the previous premises.
  4. Therefore, the person allowing it to exist isn't omnibenevolent or at least not that and capable of doing anything about it.

I think you still have some question marks to fill in. But thanks for starting to specify where you think the goalposts should be. We didn't even have that before.

I don't see what's so complicated about "ichtyosis' existence is obviously (to humans) bad (for humans) and doesn't have any visible positives (for humans), like 'necessary for the ecosystem' or whatever. Theists, why don't you show why your omnibenevolent God that allegedly prioritizes humans allows ichtyosis to exist?".

(If God doesn't prioritize humans then let the bacteria worship him, we're clearly outvoted.)

Is that argument "not coming from reality"? Where else is it coming from, then?

Because the above commenters said "cannot". They made a claim. An impossibility theorem. Only, they actually didn't say anything about how it's supposed to work. How's this impossibility theorem supposed to work?

It doesn't confirm materialism and rationalism are correct, only that the fact the Creator is omnipotent, omniscient and omnibenevolent. He can be two, but not all three.

Triple-omni God is construct of Greek philosophy and explicitly unbiblical idea.

No idea why Christians decided to adopt it as cornerstone of their theology.

I'd say that the usual theodicy formulation is extremely overstated, and in fact the observed universe is incompatible with a god that's just kinda potent, benevolent and scient. It is very easy to imagine limited supernatural powers falling well short of omni-anything which could vastly improve the world.

Unsurprisingly, just about the only sensible theodicy I've seen is Scott's Answer to Job, and that's a creative writing exercise, not an attempt to explain the world. Though he does now have a link to actual apologetics using this line of reasoning.

Sure, let's all worship the true prophet Zoroaster. It was the Orthodox Christians who took this to far.

Hell, even before(and after) Zoroaster, who the Judeans and then Christians stole so much of their theology and cosmology, various pagan and animistic faiths allowed for a world constructed by fickle gods and spirits, not much different from men, except in form and grandeur.

Not in the least. My beef is with the claim of omnibenevolence when reality demonstrates that is categorically untrue.

It doesn't even align with such an entity at least being kind, what exactly did a newborn infant do to deserve it?

The universe aligns more with utter apathy than either love or hatred.

If you end up in a situation where you are initialized with a prior of 1 in the existence of God/The Truth of Christianity, then you are perfectly rational in assuming all observed evidence in that light and then assuming any discrepancies or paradoxes arise from an incomplete understanding of the world.

If your prior is set at .75%, or .6%, or .50000001%, what changes? You still have to make sense of an incomprehensibly vast dataset based on limited information. You still have to assign bespoke weights to any given piece of evidence, based on bespoke assessments of value and mangitude.

The reason 1 and 0 are bad ideas because of Bayesian updates work, 1 implies infinite certainty, and 0 implies infinite disbelief. When you contend with a merely finite amount of evidence, no matter how vast it is, it moves the needle not one inch.

For any other values, at least in theory, a sufficient amount of evidence makes two entities with different priors converge. If they're ideal Bayesian agents, then they can just talk to each other and end up at the same spot by the Aumann Agreement Theorem, not that I can particularly follow the latter myself. The former is robustly true.

As for how that works in human neurology? No clue. But we know that LLMs initialized from random weights manage it, something to do with them being biased for simplicity in their world models, and that seems to work in practise. Presumably humans do something similar too, since we have some mechanism for teasing out useful ontologies, and it doesn't have to be particularly complex, since random noise can be coaxed there with simple algorithms.

...Sorry, I didn't mean to post that when I did, but since you replied, I'll undelete it.

The reason 1 and 0 are bad ideas because of Bayesian updates work, 1 implies infinite certainty, and 0 implies infinite disbelief.

Another way to phrase this would be that 1 and 0 represent bedrock commitment to an axiom or principle. What should my prior be that it's a good idea to cheat on my wife, or that committing murder for sport is a bad idea? Do you think sufficient evidence exists to preclude decimal notation for either question? Do you think it is beneficial to leave these as routinely-updated open questions throughout one's life?

It seems to me that there are no shortage of instances where, even if the evidence is insufficient to support a prior anywhere near 0 or 1 on a question, life requires us to abandon the decimals in favor of one integer or the other. Would you disagree?

Put another way, there is a difference between questions of what is, and questions of what we will do. Certainty is a detriment to the former and a benefit to the later, and it is simply not the case that either of the two are optional. Ideally, one lowers one's priors when figuring out the truth of things, makes a decision, and then raises one's priors when executing that decision, with the lowering and the raising being proportional to the importance of the question. But this is the problem: how do you distinguish people who've committed to a decision arrived at with imperfect knowledge from people who are incapable of examining the nature of the decision in the first place?

For any other values, at least in theory, a sufficient amount of evidence makes two entities with different priors converge.

In real life, is this effect strong enough to allow you to differentiate between a prior set at 1 before evidence, versus a prior arriving at .50001 after evidence? That is to say, do you believe you can rigorously differentiate between people who are not engaging in good faith, versus people who are engaging in good faith but are not convinced by your arguments?

The former is robustly true.

...And entirely irrelevant to actual human behavior, no? While evidence should do this, it very, very rarely does so, and you should be able to observe from the times that it has not done so for you yourself that the problem is not just people ignoring evidence, but that the effect, while real, is too weak to overmatch other influences in any but the most advantageous circumstances. Human reason is not deterministic in any practical sense, and this fact can be readily discerned by anyone willing to engage in a modest amount of introspection.

It seems to me that framing disagreement in terms of strength of priors in decimal notation, like Utilitarian calculus, is an attempt to quantify the unquantifiable. I don't believe you or anyone else can rigorously assign a prior to even the first decimal place for a real-world question of appreciable complexity, like, say, who the next president is going to be.

Presumably humans do something similar too, since we have some mechanism for teasing out useful ontologies, and it doesn't have to be particularly complex, since random noise can be coaxed there with simple algorithms.

What useful predictions does this theory allow you to make?

What should my prior be that it's a good idea to cheat on my wife, or that committing murder for sport is a bad idea? Do you think sufficient evidence exists to preclude decimal notation for either question? Do you think it is beneficial to leave these as routinely-updated open questions throughout one's life?

People regularly cheat on their wives, and at least some people commit murder for the thrill of it, leaving aside if it's a good idea.

While it would be pretentious to put too many decimal points on any reply here, I can confidently say that the "carefully" considered answer to both are sub 1%, to the point that the expected value of information from crunching it more explicitly is lackluster to say the least. Of course, I have little intrinsic desire to do each, if someone cared more they should spend more time considering it.

In real life, is this effect strong enough to allow you to differentiate between a prior set at 1 before evidence, versus a prior arriving at .50001 after evidence?

Yes? I mean, can you claim you've literally never seen anyone change their mind? People can and do express confidence levels, and there has been considerably research that shows that when someone says something is "very likely" or "unlikely", that correlates to a numerical probability even if it's noisy.

If I submit someone convincing evidence and they don't even deign to consider it or, or outright admit nothing can sway them, then I have no qualms about declaring them zealots who are immune to evidence against their claims. This is not the same as normative claims which can't be represented by priors, just utility functions, but religion as practised or preach isn't just normative, even if I think the norms in question are stupid.

I've had one person here, @Meriadoc, admit that there's nothing I or anyone, even with a trillion dollars in funding, including more funding to counter-investigate our claims, can do to change his mind regarding the truth of Mormonism. If that's not a stuck prior at 1, I don't know what is!

I don't believe you or anyone else can rigorously assign a prior to even the first decimal place for a real-world question of appreciable complexity, like, say, who the next president is going to be.

You don't see me doing that, at least unless the magnitude of the prior is significantly smaller than 1%, and even then I consciously refrain from dumb statements like claiming I only think there's a 0.0000000001% chance something is true, because the odds of me simply being insane are higher.

I usually express myself along the lines of ~10% margins, and I have empirical evidence that I am a well calibrated person, the last time I took a test that explicitly tested it, it was almost perfect from the entire range of 0-100%, in 10% buckets. So I can easily say that I'm 50-70% sure of a proposition and know that means something, and I don't claim I'm 55.67% sure.

What useful predictions does this theory allow you to make?

A newborn baby, confronted with a theoretically infinite number of explanations and world models for all the enormous amounts of evidence they're bombarded with, almost always swiftly and inevitably develops complex ontologies like object permanence, and even an understanding of the classical laws of motion.

You've previously claimed that the sheer difficulty of grappling with near infinite ways of interpreting any given sample of information precludes us from being certain about anything, but babies and the simple algorithms underling LLMs such as GPT-4 alike don't give a shit, if there's signal in the noise they'll find it.

The Simplicity Prior, or an applied form of Occam's Razor, works so well in practice that I have no qualms about relying on it.

I've had one person here, @Meriadoc, admit that there's nothing I or anyone, even with a trillion dollars in funding, including more funding to counter-investigate our claims, can do to change his mind regarding the truth of Mormonism. If that's not a stuck prior at 1, I don't know what is!

Here's what I said:

I don't love the framing of this question. If an organization with billions of dollars devoted all its resources to persuading me of something, I would be very reticent to actually be persuaded, even if the evidence seemed convincing. Presumably it could find or manufacture very good evidence for either side. Besides, people and organizations can't always just produce evidence, especially evidence which may be of a deeply personal nature.

To be very clear, I'm not yet 100% sure that my church is true.

You didn't mention any funding for counter-investigation; if you had I probably would answer differently.

Hmm, I could have sworn I did somewhere sometime, but maybe it was all in my head. Either way, apologies for being incorrect here.

reality demonstrates

What part of reality demonstrates it? Existence of ichthyosis vulgaris is a brute fact, sure. But then what? What is the next piece of reasoning? Reality provides some standard or something that somehow says that this brute fact has some implication? How are you getting to your conclusion?

That the person allowing it to exist isn't omnibenevolent or at least not that and capable of doing anything about it?

That is such a profoundly obvious takeaway I don't see how you miss it. It it's genuinely new to you, it's the problem of theodicy if one wants to use complex terminology for no good reason.

That the person allowing it to exist isn't omnibenevolent or at least not that and capable of doing anything about it?

How does that follow from something in reality? Like, fill in the details for me.

  1. Ichthyosis vulgaris exists.
  2. ??? [something coming from reality]
  3. Therefore, the person allowing it to exist isn't omnibenevolent or at least not that and capable of doing anything about it.

Ichythosis Bad.*

God Good. God Strong.

How God Let Bad Thing Happen? Ugg no understand. Why other Uggs claim God Good as well as Strong If God No Help Innocent Baby?

*citation not needed

If my concern is genuinely new to you, it's the problem of non sequitur if one wants to use complex terminology for no good reason. But yeah, sorry Ugg. We have developed thousands of years of knowledge in how reasoning works, and your argument simply isn't persuasive. It's hardly even an argument. It's just mood affiliation. Ugg's going to have to catch up with the times and figure out how to make a proper argument.

https://en.wikipedia.org/wiki/Theodicy

Not only is it an argument, it's a valid one that's been grappled with by theologians for millennia. I think it's a waste of the time in the context of the Abrahamic God, who can't be blamed for it by the simple logic that it doesn't exist.

As for whether it's persuasive to you, I can only shrug, and say you're potentially being obtuse or simply can't follow a clear argument.

More comments